Authoritarian Party SystemsParty Politics in Autocratic Regimes PDF

6"×9" b4407 Authoritarian Party Systems
Authoritarian Party Systems

Party Politics in Autocratic Regimes, 1945–2019
After the gradual slowing down of the “third wave of democratization,” electoral
authoritarianism is rapidly emerging as a dominant form of contemporary autoc-
racy. Political parties play a key role within the political and institutional structures
of electoral autocracies. Pro-regime parties provide the dictatorial executive with
electoral and legislative tools of sustaining power. At the same time, permitted oppo-
sition parties, while normally incapable of challenging the regime, are important for
regime sustainability because they perform such vital functions as co-opting actual
or potential opposition groups, facilitating power sharing, and mobilizing electoral
participation. The interactions among the dominant parties and the permitted
opposition parties, if displaying sustainable cross-temporal patterns, constitute
authoritarian party systems.
Authoritarian Party Systems provides a theoretical discussion of electoral
authoritarianism with special reference to authoritarian party systems; a method-
ological overview of party system research with special reference to the problems
caused by the authoritarian nature of the observed party systems; a comprehensive
cross-regional and historical overview of authoritarian party systems; a quantitative
analysis of their structural characteristics, including fragmentation, party system
format, volatility, and nationalization; and in-depth discussions of the political
regime determinants of authoritarian party systems and of the interplay between
party systems and other components of the authoritarian institutional order.
Quantitative analysis has been performed on an original database comprising cases
of party-structured authoritarian regimes between 1945–2019. This content of the
book is illustrated by case studies drawn from across the spectrum of contemporary
authoritarian regimes.
Grigorii V Golosov is Professor of Comparative Politics and Head of Political

Science Department at the European University at St Petersburg, Russia. He is best
known for his work on political parties, elections and electoral systems, and on
Russian national and sub-national politics. He wrote Political Parties in the Regions
of Russia: Democracy Unclaimed and several other books, and has published exten-
sively in journals, including Comparative Political Studies, Europe-Asia Studies,
International Political Science Review, Party Politics, Political Studies, Post-Soviet
Affairs, and many others. He sits on the editorial boards of several international
journals, including Party Politics, Problems of Post-Communism, REGION, Russian
Politics, and Slavic Review, and serves as an Associate International Editor of
Europe-Asia Studies. His political commentary is often quoted in the national and
international media.
b4407_FM.indd 1 1/6/2022 9:16:43 AM

B1948 Governing Asia
This page intentionally left blank
B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Q0330_9781800611160_tp.indd 1 23/12/21 3:49 PM
B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Q0330_9781800611160_tp.indd 2 23/12/21 3:49 PM
Published by
World Scientific Publishing Europe Ltd.
57 Shelton Street, Covent Garden, London WC2H 9HE
Head office: 5 Toh Tuck Link, Singapore 596224
USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601
Library of Congress Cataloging-in-Publication Data

Names: Golosov, Grigorii V, author.
Title: Authoritarian party systems : party politics in autocratic regimes,
1945–2019 / Grigorii V Golosov, European University at St. Petersburg, Russia.
Description: Hackensack, NJ : World Scientific, 2022. |
Includes bibliographical references and index.
Identifiers: LCCN 2021035023 | ISBN 9781800611160 (hardcover) |
ISBN 9781800611177 (ebook) | ISBN 9781800611184 (ebook other)
Subjects: LCSH: Dominant-party systems--Case studies. | Authoritarianism--Case studies.
Classification: LCC JF2051 .G634 2022 | DDC 324.2/1--dc23
LC record available at https://lccn.loc.gov/2021035023
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.
Copyright © 2022 by World Scientific Publishing Europe Ltd.

All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means,
electronic or mechanical, including photocopying, recording or any information storage and retrieval
system now known or to be invented, without written permission from the Publisher.
For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance
Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy
is not required from the publisher.
For any available supplementary material, please visit

https://www.worldscientific.com/worldscibooks/10.1142/Q0330#t=suppl
Desk Editors: Nimal Koliyat/Michael Beale/Shi Ying Koe
Typeset by Stallion Press

Email: enquiries@stallionpress.com
Printed in Singapore
Nimal Koliyat - Q0330 - Authoritarian Party Systems.indd 1 23/12/2021 4:09:56 pm

This book is dedicated to my beloved wife, Iulia Shevchenko.
b4407_FM.indd 7 1/6/2022 9:16:43 AM

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Acknowledgments
It has been a long road that led me to writing this book. I would like to start
with expressing my lasting gratitude to two Oxford academics who met me at
the entry to the Western academy. One of them was Mary McAuley, then of
St Hilda’s College, who selected me for a research trip to Oxford in 1993, and
the other, Peter Pulzer of All Souls College, who, figuratively speaking, met
me at the gates. I am also indebted to my mentors at the Central European
University, particularly to Laszlo Bruszt, Nenad Dimitrijevic, and Gábor
Tóka. Other international academics who helped and inspired me are too
many to name here, but I have to particularly thank George W. Breslauer,
Kenneth Janda, and Scott Mainwaring. Among the numerous Russian schol-
ars whose help meant a lot to me, I am grateful to Mikhail Ilyin and to those
many colleagues, including Aleksander Kynev and Arkady Lyubarev, who
participated with me in the Interregional Electoral Network of Assistance
(IRENA) in 2007–2011.
My work on this project would be unthinkable without the continuous
support of my home institution, the European University at St Petersburg,
and without the unique intellectual environments created and sustained by
my colleagues in the Department of Political Science, Oleg Kharkhordin,
Ivan Kurilla, Mikhail Turchenko, Margarita Zavadskaya, and especially
Vladimir Gel’man. Together we have passed through the storms of academic
life in Russia, including two closures of the European University for polit-
ical reasons. I am indebted to the administrator of the department, Anna
Gasanova, for making it possible for me to combine the duties of the head
of the department with intensive academic writing.
ix
b4407_FM.indd 9 1/6/2022 9:16:43 AM

b4407 Authoritarian Party Systems
6"×9"
x Authoritarian Party Systems
The immediate stimulus for writing this book came from the Acquisition
Editor at World Scientific Publishing Company, Natalie Watson. I am deeply
grateful to Dr Watson for her encouragement and fruitful discussions of
the project, and to many people at the WSP, including Michael Beale and
Nimal Koliyat, who participated in the production of the book.
Some of the contents of Chapter 4 of the book overlap with my previous
publications in Party Politics. I acknowledge this with deep gratitude to the
journal that has served not only as an outlet for my own work but also as a
continuous source of ideas and inspiration.
The Effective Number of Parties: A New Approach. The final, definitive
version of this paper has been published in Party Politics, Vol. 16, No. 2,
March 2010, by SAGE Publications Ltd, All rights reserved.
Party System Classification: A Methodological Inquiry. The final, defin-
itive version of this paper has been published in Party Politics, Vol. 17, No. 5,
September 2011, by SAGE Publications Ltd, All rights reserved.
Party System Nationalization: The Problems of Measurement with
an Application to Federal States. The final, definitive version of this paper
has been published in Party Politics, Vol. 22, No. 3, May 2016, by SAGE
Publications Ltd, All rights reserved.
b4407_FM.indd 10 1/6/2022 9:16:43 AM

Contents
Acknowledgments ix
List of Figures xiii
List of Tables xv
Chapter 1 Introduction1
Chapter 2 Defining Authoritarian Party Systems 7

Authoritarianism: Substantive Considerations 7
Authoritarianism: Operational Definition 14
Party Systems 17
Chapter 3 The World of Authoritarian Party Systems 27

Historical Antecedents 30
Early Contemporary Cases 33
The Authoritarian Party Systems of the Western Hemisphere 35
The Authoritarian Party Systems of Asia 38
The Authoritarian Party Systems of Africa 42
The Authoritarian Party Systems of
Post-Communist Countries 47
Chapter 4 The Properties of Authoritarian Party Systems 51

Party System Fragmentation 53
Party System Format 66
Party System Volatility 76
Party System Nationalization 80
xi
b4407_FM.indd 11 1/6/2022 9:16:43 AM

6"×9"
xii Authoritarian Party Systems
Chapter 5 Political Regimes and Authoritarian Party Systems 89

Authoritarian Political Regimes and Regime Genomes 90
Assigning Party Systems to Regime Genome Categories 94
Mode of Power Acquisition 102
Party Legacies 108
Empirical Methods 116
Empirical Findings 120
Chapter 6 Authoritarian Party Systems, Executive–Legislative

Relations, and Federalism 133
Types of Executive–Legislative Relations 135
Executive–Legislative Relations and the Properties
of Authoritarian Party Systems 139
Federalism and the Properties of Authoritarian Party Systems 160
Chapter 7 Authoritarian Party Systems and Electoral Systems 169

Electoral Systems and Their Effects 170
Electoral Systems and the Properties of Authoritarian
Party Systems 175
Empirical Findings 186
Chapter 8 The Destinations of Authoritarian Party Systems 195

The Legacies of Authoritarian Party Systems 195
The Present and Future of Authoritarian Party Systems 205
Conclusion 217
List of the Names of Parties and Political Organizations

in Native Languages 225
References 229
Index 271
b4407_FM.indd 12 1/6/2022 9:16:43 AM

List of Figures
Figure 3.1. The historical dynamics of authoritarian party systems:

Numbers of country cases in existence by year,
1945–201929
Figure 4.1. The segmented relative-size triangle with some data
points of significance 71
Figure 4.2. The graphic representation of party constellations in
democracies and authoritarian regimes. Plate 1: 148
democratic elections, 1987–2014. Plate 2: 294
authoritarian elections, 1945–2019. Plate 3: 57
authoritarian party systems, 1945–2019 75
Figure 5.1. The graphic representations of party constellations
in authoritarian elections by political regime genomes.
Plate 1: Monarchy. Plate 2: Military regime.
Plate 3: Single-party regime. Plate 4: Post-democracy.
Plate 5: Party electoral authoritarian regime.
Plate 6: Non-party electoral authoritarian regime 126
Figure 6.1. The graphic representations of party constellations in
authoritarian elections by institutional design types.
Plate 1: Parliamentary systems. Plate 2: Systems with
parliamentary elected executive presidents. Plate 3:
Presidential systems. Plate 4: Semi-presidential systems 155
in federal and non-federal autocracies. Plate 1:
Federations. Plate 2: Non-federations 166
xiii
b4407_FM.indd 13 1/6/2022 9:16:43 AM

6"×9"
xiv Authoritarian Party Systems

authoritarian elections by electoral systems. Plate 1:
Majoritarian systems. Plate 2: Proportional
representation systems. Plate 3: Mixed systems 190
Figure 8.1. Mean and median fragmentation of authoritarian
party systems by year, 1949–2019. Plate 1: Mean.
Plate 2: Median 207
b4407_FM.indd 14 1/6/2022 9:16:43 AM

List of Tables
Table 2.1. Authoritarian party systems of the world, 1945–2019 24

Table 3.1. The geographical and chronological distribution of
authoritarian party systems 28
Table 4.1. The values of different measures of fragmentation
for eight hypothetical vote or seat constellations 60
Table 4.2. Fragmentation of democratic and authoritarian
party s ystems (effective number of parties) 64
Table 4.3. Summary of ANOVA comparison of party systems
by fragmentation 65
Table 4.4. Bonferroni corrected comparisons of democratic
and authoritarian party systems by fragmentation 65
Table 4.5. The segments of the relative-size triangle defined
by the sizes of three largest components 72
Table 4.6. 12 hypothetical party constellations at the segment
centroids of the relative-size triangle diagram 74
Table 4.7. Volatility of democratic and authoritarian party systems 78
by volatility 78
Table 4.9. Bonferroni corrected comparisons of democratic and
authoritarian party systems by volatility 79
Table 4.10. Nationalization of democratic and authoritarian
party s ystems (index of party system nationalization) 86
by nationalization 86
xv
b4407_FM.indd 15 1/6/2022 9:16:43 AM

6"×9"
xvi Authoritarian Party Systems
Table 4.12. Bonferroni corrected comparisons of democratic

and authoritarian party systems by nationalization 86
Table 5.1. The classification of authoritarian party systems
by political regime genomes 100
Table 5.2. Index of electoral democracy by political regime
genomes100
Table 5.3. Summary of ANOVA comparison of political regime
genomes by the index of electoral democracy 101
Table 5.4. Bonferroni corrected comparisons of political regime
genomes by the index of electoral democracy 101
Table 5.5. The descriptive characteristics of the control variables 120
Table 5.6. Party system fragmentation by regime genomes 121
genomes by fragmentation 121
genomes by fragmentation 122
Table 5.9. Linear regression results, effective number of parties
by political regime genomes (n = 294 adjusted for
57 clusters; unstandardized coefficients;
clustered-robust errors in parentheses) 123
Table 5.10. Multilevel mixed-effects regression results, effective
number of parties by political regime genomes 124
Table 5.11. Two-way analysis of variance: regime genomes
and party system formats 127
Table 5.12. Party system volatility by regime genomes 128
genomes by volatility 128
genomes by volatility 129
Table 5.15. Linear regression results, volatility by political regime
genomes (n = 237 adjusted for 57 clusters;
unstandardized coefficients; clustered-robust errors in
parentheses)130
b4407_FM.indd 16 1/6/2022 9:16:43 AM

List of Tables xvii
Table 5.16. Multilevel mixed-effects regression results, volatility

by political regime genomes 131
Table 5.17. Party system nationalization by regime genomes 132
Table 5.18. Pairwise correlations (Pearson’s r), party system
nationalization by political regime genomes, n = 42 132
Table 6.1. The classification of authoritarian party systems by
institutional types 140
Table 6.2. Two-way analysis of variance: models of executive–
legislative relations and political regime genomes 141
Table 6.3. Party system fragmentation by models of executive–
legislative relations 152
Table 6.4. Summary of ANOVA comparison of institutional
design types by fragmentation 152
Table 6.5. Bonferroni corrected comparisons of institutional
design types by fragmentation 153
Table 6.6. Linear regression results, effective number of parties
by institutional design types (n = 294 adjusted for
57 clusters; unstandardized coefficients;
clustered-robust errors in parentheses) 154
number of parties by models of executive–legislative
relations154
Table 6.8. Two-way analysis of variance: models of executive–
legislative relations and party system formats 156
Table 6.9. Party system volatility by models of executive–
legislative relations 157
Table 6.10. Summary of ANOVA comparison of institutional
design types by volatility 157
Table 6.11. Bonferroni corrected comparisons of institutional
design types by volatility 158
Table 6.12. Linear regression results, volatility by institutional design
types (n = 294 adjusted for 57 clusters; unstandardized
coefficients; clustered-robust errors in parentheses) 158
b4407_FM.indd 17 1/6/2022 9:16:43 AM

6"×9"
xviii Authoritarian Party Systems
Table 6.13. Multilevel mixed-effects regression results, volatility

by models of executive–legislative relations 159
Table 6.14. Party system nationalization by models of
executive–legislative relations 159
Table 6.15. Pairwise correlations (Pearson’s), nationalization by
institutional design types, n = 42 159
Table 6.16. Party system parameters in relation to federalism 164
Table 6.17. Summary of ANOVA comparison of federations and
non-federations by fragmentation 164
non-federations by volatility 164
Table 6.19. Linear regression results, fragmentation and volatility
by federalism (n = 294 adjusted for 57 clusters;
unstandardized coefficients; clustered-robust errors
in parentheses) 165
Table 6.20. Multilevel mixed-effects regression results,
fragmentation and volatility by federalism 165
Table 6.21. Two-way analysis of variance, federalism, and
party system formats 166
Table 7.1. Electoral systems used in the countries with
authoritarian party systems 176
Table 7.2. Party system parameters in relation to electoral systems 187
Table 7.3. Summary of ANOVA comparison of countries
with different electoral systems by fragmentation 187
Table 7.4. Bonferroni corrected comparisons of countries
with different electoral systems by fragmentation 188
Table 7.5. Linear regression results, fragmentation by electoral
systems (n = 294 adjusted for 57 clusters;
unstandardized coefficients; clustered-robust
errors in parentheses) 188
Table 7.6. Multilevel mixed-effects regression results, party
system fragmentation by electoral systems 189
b4407_FM.indd 18 1/6/2022 9:16:43 AM

List of Tables xix
Table 7.7. Two-way analysis of variance: electoral systems

and party system formats 191
Table 7.8. Summary of ANOVA comparison of countries
with different electoral systems by volatility 191
Table 7.9. Bonferroni corrected comparisons of countries
with different electoral systems by volatility 192
Table 7.10. Linear regression results, volatility by electoral
systems (n = 294 adjusted for 57 clusters;
unstandardized coefficients; clustered-robust
errors in parentheses) 192
Table 7.11. Multilevel mixed-effects regression results,
volatility by electoral systems 193
nationalization by electoral systems, n = 42 193
Table 8.1. Reasons for termination of the political regimes
conjoined with the extinct party systems 199
Table 8.2. The descriptive characteristics of non-dichotomous
variables in the analysis of post-authoritarian
party system change (n = 34) 202
Table 8.3. Linear regression results, factors of authoritarian
party s ystem change measured as post-authoritarian
volatility (n = 34, numbers of observations
in parentheses) 203
Table 8.4. The properties of authoritarian party systems
by survival status, descriptive statistics 205
Table 8.5. Distribution of the extinct and existing authoritarian
party systems by political regime genomes and
institutional design types (%) 208
Table 8.6. The emerging authoritarian party systems as of the
end of 2019 211
b4407_FM.indd 19 1/6/2022 9:16:43 AM

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Chapter 1
Introduction
A
bout 30 years ago, Valerie Bunce’s (1999) seminal book opened up
an unusual perspective on authoritarian institutions. The book, tell-
ingly titled Subversive Institutions, provided a convincing account of
a crucial role played by the constitutional arrangements inherited from the
Soviet era — primarily federalism, but also representative assemblies — in
the demolition of the Communist political regime in the Soviet Union. This
account departed from a long-standing tradition of treating authoritarian
institutions as mere formalities devoid of any substantive importance or, as
Gandhi and Przeworski (2007) put it when characterizing this traditional
approach, “window-dressing.” Indeed, in the classic treatments of Soviet
politics, these institutions were often discarded as lacking any tangible effects
at all (Friedrich and Brzezinski, 1961) or, at best, playing a secondary role in
providing a propaganda topic for domestic and international consumption
(Fainsod, 1958).
Contemporary scholarship on authoritarian institutions breaks drasti-
cally from the “window-dressing” approach to authoritarian institutions. The
main reason for this change of perspective is thus. The scholars of the past
dealt mostly with the straightforwardly dictatorial regimes and their often
idiosyncratic institutional arrangements. Contemporary autocracies are
different in that they widely use institutions normally associated with democ-
racy, including partially competitive elections, legislatures, and political
parties (Magaloni and Kricheli, 2010). While the role of “window-dressing”
is still performed by authoritarian institutions (Joseph, 1997), this role is now
conceived as auxiliary to more important functions. In a drastic departure
b4407_Ch01.indd 1 1/6/2022 9:14:38 AM

6"×9"
2 Authoritarian Party Systems
from the “subversive institutions” argument, the currently most influential

approach to authoritarian institutions, as initially proposed by Huntington
(1968) and developed by Geddes (2003), views their primary function not in
subverting autocracy but rather in making it more sustainable and durable.
As pithily put by Slater (2003, p. 82), authoritarian institutions “primarily
serve to provide a stable basis for domination.”
This growing stream of literature identifies many reasons why author-
itarian institutions are important. Several studies demonstrate that legis-
latures enhance the survival of authoritarian regimes by securing several
benefits for them, including such apparently apolitical ones as higher rates
of economic growth and foreign investment (Gandhi, 2008; Wright, 2008;
Boix and Svolik, 2013; Jensen et al., 2014). The functions of authoritarian
elections are more narrowly political, but in this capacity, they are essential
for sustaining contemporary dictatorships. Elections help to solve intra-
regime conflicts that would otherwise have destabilizing effects (Geddes
et al., 2018), thus playing a central part of the survival strategies of the auto-
crats (Magaloni, 2006). In particular, elections serve as a counterbalance to
the military or factions within the military that can pose existential threats
to the incumbent authorities.
From a more general perspective, elections facilitate power sharing
and cooptation of the potential or actual adversary groups within the elite.
Elections make it possible for the autocrat to give voice to these groups,
bargain with them, and make policy concessions to address their demands
(Gandhi and Przeworski, 2006, p. 17; see also Frantz and Kendall-Taylor,
2014). At the same time, elections are used to distribute economic transfers
and rents (Wintrobe, 1998; Bueno de Mesquita et al., 2003), thus appeasing
the potential opponents of the regime. In this way, cooptation mitigates
threats to autocracies and increases their longevity (Gandhi, 2008; Gandhi
and Przeworski, 2007; Brancati, 2014). To a greater extent than public opin-
ion polls, elections are useful for collecting information about the political
attitudes of the population (Simpser, 2013; Miller, 2015b), the actual scope
of influence enjoyed by the opposition (Blaydes, 2010), and the regime’s
b4407_Ch01.indd 2 1/6/2022 9:14:38 AM

Introduction 3
mobilization capacity (Cheibub and Hays, 2017). Authoritarian elections

serve as an important arena of pro-government political mobilization, both
by intensifying propaganda efforts (Hellmeier and Weidmann, 2020) and
by buying support from political elites and citizens through clientelistic
exchange (Svolik, 2012).
While some of the literature cited above does not draw a distinction
between non-competitive elections and those elections in which the partic-
ipation of opposition parties is permitted, there is a broad scholarly consen-
sus regarding the importance of limited inter-party electoral competition
under authoritarianism (Brownlee, 2007). In fact, all arguments in support
of the importance of authoritarian institutions can be specified in a way
that emphasizes the role of multiparty politics. Most of these arguments
have been originally formulated or developed with reference to the role
of inter-party competition in power sharing, cooptation, and/or political
mobilization. Given that the “window-dressing” element of authoritarian
institutionalism has not withered away — it is still important for the autocrats
to emulate democracy as closely as they can afford without putting their
survival in power into danger (Lührmann et al., 2018) — it would be fair
to posit that more than ever before, the autocracies of today are disposed
toward party politics. The opening thesis of Richard Katz’s (1980, p. 1)
influential book states that “modern democracy is party democracy.” Today,
it can be said without much reservation that “modern authoritarianism is
party authoritarianism.”
Indeed, recent scholarship on authoritarianism has paid close attention
to political parties. On the one hand, there is an already large and contin-
uously growing body of literature on the dominant pro-regime parties that
have emerged in a majority of contemporary authoritarian regimes and
serve as important pillars of autocracy. These parties are indeed pivotal
in a majority of authoritarian party settings, and their activities are very
illuminating in regard to understanding the basic mode of operation of
contemporary non-democratic regimes (Magaloni, 2006; Brownlee, 2007;
Greene, 2007; Blaydes, 2010; Schedler, 2013; Simpser, 2013; Reuter, 2017).
b4407_Ch01.indd 3 1/6/2022 9:14:38 AM

6"×9"
On the other hand, there is a growing scholarly interest in the activities of

the opposition parties, mainly from the angle of their ability to effectively
challenge the authoritarian rule (Howard and Roessler, 2006; Bunce and
Wolchik, 2010; Wahman, 2011; Donno, 2013; Gandhi and Reuter, 2013;
Gandhi and Ong, 2019; Ufen, 2020). While much of this scholarship does
establish conjunctions between the two components of authoritarian party
politics, the government and the opposition, from the point of view of
strategies pursued by the political actors and the resources at their disposal,
little research has been done on the structural properties of inter-party
competition under authoritarianism.
The structural properties of party politics are captured in the concept of
the party system, commonly defined as “the system of interactions resulting
from inter-party competition” (Sartori, 1976, p. 44). These properties are sys-
temic. They cannot be reduced to the properties of the component elements
of the system, that is, individual parties or their clusters. The literature on
democratic party systems is voluminous (for a recent overview, see Siaroff,
2019), and their properties, while having been closely examined in the past
literature, remain focal in the ongoing research. It would be fair to say that
to a large extent, democratic party politics are studied at the level of party
systems, not individual parties. This research focus can be justified with
reference to important substantive considerations, such as that “it is not the
institutionalization of political parties but the institutionalization of party
systems as a whole that has fostered the prospects for democratic survival
in Europe” (Casal Bértoa, 2017, p. 402).
However, a systemic focus has been largely avoided in research on
authoritarian party politics. Insofar as the structural properties of author-
itarian party systems are addressed at all, they are often presumed to be
simple and unproblematic, which is best expressed in an almost universally
held assumption that authoritarian party systems necessarily include fewer
important parties than it is common in democracies. Such assumptions
may be correct with regard to general tendencies, but, even if they are abso-
lutely plausible, they cannot be accepted without proper empirical testing.
b4407_Ch01.indd 4 1/6/2022 9:14:38 AM

Introduction 5
In particular, it is important to establish the conditions under which such

assumptions actually hold. This approach is essential in social sciences
because all social laws are qualified ceteris paribus (Kincaid, 1990).
The purpose of this cross-national inquiry is to fill this lacuna in the
literature by taking a step toward identifying and explaining the properties
of authoritarian party systems. Much attention is paid to the political regime
determinants of the structural properties of authoritarian party systems, on
the one hand, and to the impact exerted by institutional factors, on the other
hand. These factors, including executive–legislative relations, federalism, and
electoral systems, are generally recognized as important determinants of
party system properties in democracies, and within the agenda of research on
authoritarian institutionalism, assessing their impact upon party systems is
no less important. Chronologically, the study is restricted to the post-World
War II period up to 2019, even though some attention is also paid to the
earlier instances of authoritarian party systems.
The structure of the book is related to this purpose. Chapter 2 defines
authoritarian party systems and provides a general theoretical discussion
of electoral authoritarianism in relation to party politics. On this basis, the
universe of authoritarian party systems is delineated as a set of empirically
observable cases. Chapter 3 provides information about these cases, as well
as about the historical antecedents of authoritarian party systems. This
largely descriptive chapter is deemed to be necessary not only because any
phenomenon should be considered in a historical retrospective, but also
for a more practical reason of shedding some light on those authoritarian
party systems that have received little attention in the literature and become
forgotten over time. But they still constitute important cases for cross-
national empirical analysis.
Chapter 4 provides a discussion of the existing methods of measuring
main party system properties: fragmentation, format, volatility, and nation-
alization. These methods are addressed in the context of studying authori-
tarian party systems. The methodological focus of this section allows for a
structured comparison of authoritarian party systems with the party systems
b4407_Ch01.indd 5 1/6/2022 9:14:38 AM

6"×9"
of democracies, both well-established and emerging. Chapter 5 explains the

observed properties of authoritarian party systems by relating these prop-
erties to the characteristics of different varieties of authoritarianism. This
chapter introduces and employs the concept of political regime genome,
understood as an integral characteristic of the regime’s origin and mode of
operation, as an explanatory tool in research on authoritarian party systems.
Chapter 6 discusses theoretically and explores empirically the impact
of major institutional arrangements of horizontal (executive–legislative
relations) and vertical (federalism) separation of powers upon authoritarian
party systems. Chapter 7 follows the same route when focusing on the effects
of electoral rules. Chapter 8 discusses the destinations of authoritarian party
systems from two different substantive perspectives: first, it analyzes the
legacies of authoritarian party systems in those countries where they have
ceased to exist; and second, it provides a brief discussion of the remaining
authoritarian party systems, which leads to theoretically and empirically
informed speculation regarding the prospects of party politics under
authoritarianism.
b4407_Ch01.indd 6 1/6/2022 9:14:38 AM

Chapter 2
Defining Authoritarian
Party Systems
I
n this book, authoritarian party systems are defined as party systems
that exist under the conditions of authoritarianism. This definition,
simplistic as it may seem, departs from some of the approaches to
authoritarian party systems that are widespread in the contemporary liter-
ature. On the one hand, authoritarian party systems are often considered as
practically equivalent to the constellations of political forces characterized by
single-party dominance. On the other hand, the very existence of political
parties, irrespective of their ability to compete in elections, is often consid-
ered as providing sufficient grounds for recognizing that an authoritarian
party system is in place. The approach of this book is different in that it
neither considers single-party dominance, however defined, as the only
possible form of the existence of authoritarian party systems, nor ascribes
to this category such situations in which parties do not compete for votes.
The goal of this chapter is to explain the theoretical reasons for this approach
and, on this basis, to define authoritarian party systems in operational terms.
Authoritarianism: Substantive Considerations

From the brief preliminary definition provided at the beginning of this
chapter, it can be inferred that a more detailed definition can be obtained
at the intersection of two concepts, those of party systems and authoritar-
ianism. I start with the latter. Authoritarianism is normally juxtaposed to
democracy as a political regime. Based on the well-known definitions of
7
b4407_Ch02.indd 7 1/6/2022 9:14:50 AM

6"×9"
Fishman (1990), Geddes (1999), and O’Donnell (2004) and taking into
account more recent theoretical contributions (Svolik, 2012; Geddes et al.,
2014), a political regime can be defined as a set of rules, either formal or
informal, that are essential for selecting political leaders, for maintaining
them in power, and therefore, for the succession of power. By extension, these
three components of the political regime often determine how the policies
of the political leadership are selected and implemented, even though it
has to be immediately recognized that in many domains, specific policies
can be selected and implemented irrespective of political regime properties
(Adamson, 2020).
While it is common to divide the universe of political regimes into two
broad categories, democratic and authoritarian regimes, supplementing this
simple dichotomy with different intermediate types and/or subtypes is by
no means unusual (Bogaards, 2009). But the categories of democracy and
authoritarianism are present in all influential classifications (Kailitz, 2013),
which indicates that drawing a theoretical distinction between them is a
primary condition for further analysis. Proceeding from the above defini-
tion of a political regime, democracy is a regime in which political leaders
are selected and maintain their power by popular will, as expressed in free
and fair elections.
It is useful to clarify this definition by quoting Przeworski (1991, p. 11):
democracy is a “system in which parties lose elections.” This clarification
allows for avoiding the thorny question of the prerequisites for free and
fair elections. Many of these prerequisites are technical and, while highly
desirable, are effectively absent in some of the well-established democra-
cies (Goodwin-Gill, 2006). Even electoral fraud is not entirely unfamiliar
to some of the undeniably democratic countries (Hill et al., 2017). But if
elections are so organized and conducted that they cannot be lost by the
incumbent ruler, democracy is not in place because popular will cannot be
properly exercised. This implies that under authoritarianism, political power
is obtained and maintained by using mechanisms that are different from
elections, irrespective of whether they are conducted or not.
b4407_Ch02.indd 8 1/6/2022 9:14:50 AM

Defining Authoritarian Party Systems 9
While the real-life manifestations of democracy vary depending on the

national cultural and historical backgrounds, all these manifestations can be
described and analyzed as falling under a single category. Different as they
are, all democracies are democracies as long as they satisfy the criterion of the
decisiveness of popular will in terms of power allocation and preservation.
Distinctions within this category are normally established by estimating the
degree of presence of certain conditions that are desirable but not necessary
for the free expression of popular will, which allows for differentiating full-
scale liberal democracies from imperfect electoral democracies (Møller and
Skaaning, 2013). At the same time, the sub-categories can be identified in
accordance with the formal rules that define the distribution of power within
the overall political structure. This approach allows for making a distinction
among parliamentary, presidential, and intermediate varieties of democratic
institutional design (Shugart and Carey, 1992).
It has to be mentioned that some scholars view the so-called hybrid
or semi-democratic regimes as an analytical category that is either sub-
stantively separate from democracy and authoritarianism (Morlino, 2009),
which is questionable (Mufti, 2018), or can be productively employed as an
empirically-grounded residual category (Wigell, 2008; Mainwaring et al.,
2001; Gilbert and Mohseni, 2011), which is obviously acceptable in certain
research settings. For most purposes, however, it is sufficient to classify
hybrid regimes as belonging to the category of authoritarianism (Sartori,
1987; Cheibub et al., 2010), and I follow this approach.
Putting aside the sub-national level of politics where the electoral
failures of authoritarian regimes are neither unusual nor necessarily con-
sequential, is it true that the ruling parties of authoritarian regimes do not
lose national-level elections? There is a stream of literature on what is called
democratization by elections (Schedler, 2002a; Lindberg, 2009; Wahman,
2013). The wording of the term apparently suggests that the autocrats can, in
fact, lose elections, and several examples of such events are indeed available.
But, as demonstrated by Morgenbesser and Pepinsky (2019, p. 3), “elections
are almost always the culmination rather than the cause of democratization.”
b4407_Ch02.indd 9 1/6/2022 9:14:50 AM

6"×9"
In particular, elections provide a convenient route of exit from power for

those autocrats who have already lost their struggles for political survival. To
prevent a threat of mass violence or military takeover, autocrats may concede
to democracy (Acemoglu and Robinson, 2006). Yet, as long as the autocrat’s
grip on power remains strong, he is capable of winning elections by engaging
a set of devices dubbed “the menu of manipulation” by Schedler (2002b).
It would lead far beyond the scope of this inquiry to analyze the individ-
ual items on the menu of manipulation in detail. From a wide perspective, all
these items fall under the category of electoral malpractice (Schedler, 2002b;
Elklit and Reynolds, 2005; Calingaert, 2006). A useful overall framework
for understanding this phenomenon is provided by Birch (2011) who estab-
lishes several general criteria for the fairness of elections and, on this basis,
identifies main varieties of electoral malpractice: (1) manipulation of the
law by setting unfair rules of elections, (2) manipulation of the voter, which
includes the manipulation of the voters’ sincere preferences and forcing the
voter into insincere electoral behavior by vote buying or intimidation, and
(3) manipulation of the vote, i.e., outright election fraud. Fortin-Rittberger
(2014) shows that the use of these strategies of manipulation is variable across
different political regimes, but in autocracies, as demonstrated by a large
and constantly growing body of research, at least some of them are always
in place (Greene, 2007; Beaulieu and Hyde, 2009; Skovoroda and Lankina,
2017; Saikkonen and White, 2020).
In contrast to democracy, authoritarianism is not a single political
regime but a set of different political regimes that can be distinguished
from each other by the same criteria that allow for differentiating them all
from democracy: under these regimes, political leaders obtain and maintain
power in different ways, even though none of these ways can be viewed as an
expression of popular will in free and fair elections. Some of these ways are
easily empirically identifiable, which is reflected in a number of classifications
of authoritarian regimes proposed in the recent literature.
Geddes (2003) identifies three main categories: party-based regimes,
military regimes, and personalist regimes. In the classification of Hadenius
b4407_Ch02.indd 10 1/6/2022 9:14:50 AM

and Teorell (2007), the personalist regime category is omitted, and the
categories of one-party autocracies and multiparty autocracies are included.
Magaloni and Kricheli (2010) follow the same line by drawing a distinction
between single-party and dominant party regimes. They also add the category
of monarchy. While the theoretical bases for introducing these particular
categories are well articulated by the authors of these classifications, all
of them use the mode of obtaining and maintaining power as a primary
criterion. It has to be noted that by monarchies, these authors understand
political regimes in which the royal rulers possess significant effective, rather
the purely ceremonial, powers. Thus defined, the category applies neither to
liberal democracies with hereditary heads of states nor to several formally
monarchical authoritarian regimes. For instance, contemporary Cambodia,
while being a monarchy from the constitutional standpoint, is effectively a
party regime (Morgenbesser, 2019).
The classification of authoritarian regimes used in this study builds
on the existing typologies. First, I take into account the properties of the
narrow ruling group that makes decisions regarding access to power. From
this perspective, the most easily identifiable types are monarchies, where
these decisions are made by the acting monarch, his/her family, and the
royal court, normally in accordance with the rules of hereditary succes-
sion; party regimes, in which they are made by the acting party leader and
other prominent members of party leadership, often in accordance with the
intra-party regulations; and military regimes, in which they are made by the
chief military commander in power and other top military officers, often in
accordance with their relative weight within the military corporation and
sometimes on the basis of informal power-sharing agreements among the
military leaders.
The role of formal rules in determining who governs is crucial for
political regime institutionalization (Pérez-Liñán and Mainwaring,
2013). From this perspective, military regimes score particularly low
because they normally do not establish any formal rules regulating
access to power. Low institutionalization is also characteristic of those
b4407_Ch02.indd 11 1/6/2022 9:14:50 AM

6"×9"
political regimes that are based on personal domination without relying

on such mechanisms as hereditary succession, intra-party mechanisms,
or distribution of senior command positions within the military corpo-
ration. It is certainly true that all types of political regimes, including
even democracy, include personalism as a significant component. Yet,
in genuinely personalist regimes, decisions regarding access to power
are made by the acting autocrat and the “inner circle” consisting of the
autocrat’s trusted political allies. Informal power-sharing agreements
among them can play a role, but this role is insignificant even in com-
parison with military regimes.
For the reasons described in the Chapter 1, and particularly because of
the lack or insufficiency of the institutionalized mechanisms of accession
to power and succession, all authoritarian regimes can employ elections
as secondary institutions. The utility of elections for different types of
authoritarian regimes varies depending on their fundamental properties.
In monarchies, elections can be used to form representative assemblies
with limited legislative or merely consultative functions (Ihalainen et al.,
2016). The party regimes of the past, such as the communist regimes,
normally conducted elections. Even though many of these elections were
non-competitive and largely symbolic (Zaslavsky and Brym, 1978), there
were several instances of non-communist authoritarian regimes that
combined single-party dominance with limited electoral competition
(Hermet et al., 1978). Military regimes often justified their interventions in
domestic affairs by the need to restore political order, revive the economy,
and protect the nation (Nordlinger, 1977), thus emphasizing that they
did not need lasting political institutions, but some of the long-standing
military autocracies did make use of elections (Falleti, 2011; Croissant
and Kamerling, 2013).
Unlike monarchies and party autocracies, personalist dictatorships nor-
mally do not develop regime-specific institutions. It is usual for non-military
dictators to seek prolongation of their rule by winning elections that, under
b4407_Ch02.indd 12 1/6/2022 9:14:50 AM

conditions of authoritarianism, are neither free nor fair but still involve
inter-party competition. Muammar Gaddafi’s Libyan Jamahiriya (Totman
and Hardy, 2015) remains exceptional as a far-reaching attempt to create
an entirely new state organization in order to disguise the dictatorial nature
of the regime.
The rarity of self-styled institutional arrangements among contemporary
personal dictatorships makes them particularly likely to emulate institutions
that are characteristic of different regime types. In the 1960s through the
early 1980s, many personalist regimes — as well as some military dictator-
ships (Decalo, 1985b) — successfully emulated the institutional structures of
communist party regimes. The collapse of global communism in the end of
the 1980s effectively eliminated this option from the institutional toolkit
of non-communist autocracies. As a result, all contemporary personalist
dictatorships claim democratic credentials, and nearly all of them permit
certain degrees of electoral competition. It has been noted in the literature
that in order to neutralize the threat to their power posed by the military,
personalist dictatorships tend to create political parties or use political parties
inherited from the previous periods of political development (Frantz and
Kendall-Taylor, 2017).
From the perspective outlined above, the empirical field of this research
embraces all instances of party-structured competition in direct elections
that can be observed in different authoritarian regimes. Correspondingly, the
term electoral authoritarianism, as employed in this study, does not refer to a
specific category of authoritarian regimes as defined by the fundamental cri-
terion of the patterns of access to power. Rather, it is a cross-cutting category
that is relevant to all regime types as long as they involve party-structured
elections. This approach involves no preliminary assumptions about the
possible shapes of authoritarian party systems, which is why they cannot
be a priori ascribed to the category of single-party dominance. Quite the
reverse, variations of party systems across authoritarian regime types form
a focal point of this inquiry.
b4407_Ch02.indd 13 1/6/2022 9:14:50 AM

6"×9"
Authoritarianism: Operational Definition

The definition of authoritarian party systems as those that exist under the
conditions of authoritarianism clearly implies that these conditions have to
be identified empirically, which calls for an operationalization of the con-
cept. Party systems manifest themselves primarily in the form of election
results. Therefore, the first step in delineating the scope of empirical inquiry
is to identify all elections held in authoritarian conditions throughout the
period of observation, which can be achieved by using a year-by-year,
cross-national database that is aimed specifically at measuring democracy
and/or authoritarianism.
Since the period of observation in this study embraces the whole post-
World War II period up to 2019, some of the available databases that are
commonly used for similar purposes cannot be employed in this study
because their chronological scope is insufficient. This applies to such a
popular instrument as the Freedom House ratings that are available starting
from 1972. Besides, it has to be mentioned that according to some critics,
this instrument is often misplaced because it measures freedom rather than
democracy (Munck and Verkuilen, 2002).
Some of the democracy indices are presented in databases that cover
even more recent time periods, such as the Voice and Accountability data-
base (Kaufmann and Kraay, n.d.), or have not been fully or partially updated
for the 2010–2019 period (Vanhanen, 2000; Bernhard et al., 2001; Cheibub
et al., 2010; Boix et al., 2013). The Lexical index of electoral democracy
(Skaaning et al., 2015) is fully satisfactory in terms of chronological cover-
age and poses a number of other advantages, but the classification criteria
used in this dataset are very narrow, mostly confined to the very existence
of electoral competition and some of its most easily observable properties.
This approach, while consistent with the goals set by the creators of the
Lexical index, is not sufficient for this study.
Two of the available datasets, Polity IV (Marshall et al., 2016) and
V-Dem (Coppedge et al., 2011), are outstanding in their chronological and
b4407_Ch02.indd 14 1/6/2022 9:14:50 AM

geographical coverage. The Polity index is produced and systematically

updated by one of the most long-standing projects of this kind, and it has
exerted a profound and lasting impact on the past and ongoing scholarship.
As widely recognized in the literature, it is not devoid of significant short-
comings, which is best summarized by Treier and Jackman (2008) who
come to the conclusion that “skepticism as to the precision of the Polity
democracy scale is well founded” (p. 201). Boese (2019) demonstrates quite
convincingly that “the measures developed by the Varieties of Democracy
(V-Dem) project outperform Polity2 and Freedom House index with respect
to the underlying definition and measurement scale, as well as the theoretical
justification of the aggregation procedure” (p. 95).
While the discussion on democracy measurement is certainly beyond
the scope of this study, the available in-depth treatments of the topic suggest
that V-Dem is a highly useful instrument, particularly because it has been
developed relatively recently and seeks specifically to address some of the
pitfalls of the previous efforts at democracy measurement (Coppedge et al.,
2017). This explains my choice of V-Dem as a tool for distinguishing author-
itarian regimes from democracies and thereby, for case selection. A brief
description of the V-Dem database, available in detail at www.v-dem.net at
the time of writing, is in order.
As explained at the above-mentioned website, the V-Dem project
provides a multidimensional and disaggregated dataset that reflects the
complexity of the concept of democracy as a system of rule that goes beyond
the simple presence of elections. The V-Dem project distinguishes between
five high-level principles of democracy: electoral, liberal, participatory,
deliberative, and egalitarian, and collects data to measure these principles. In
this study, I use the 10th version of the dataset released in 2020 (Coppedge
et al., 2020). The dataset covers 202 countries of the world, including some
of the states that are currently not in existence, for the overall period of
1789–2019, even though the chronological scope varies among countries.
The contemporary micro states are not included. The dataset includes
independent states, but some countries are coded prior to independence.
b4407_Ch02.indd 15 1/6/2022 9:14:50 AM

6"×9"
The measures of democracy provided in the dataset are continuous. One

of these measures, the v2x_polyarchy variable on the dataset, defined by the
creators of the project as the electoral democracy index, has been chosen as
most relevant to the specific goals of this study. The question addressed by
the electoral democracy index is thus: “To what extent is the ideal of electoral
democracy in its fullest sense achieved?” The index is formed by taking the
average of, on the one hand, the weighted average of the indices measuring
the freedom of association, “clean” (fair) elections, the freedom of expression,
elected officials (referring to whether the chief executive and the legislature
are appointed through popular elections), and suffrage, and, on the other
hand, the five-way multiplicative interaction between those indices. Thus,
the index allows for measuring electoral democracy, and thereby electoral
authoritarianism, both concisely and comprehensively.
The electoral democracy index is used in this study in two ways. In
its capacity of a continuous measure, it is indispensable in performing the
important task of controlling for the level of democracy in all statistical anal-
yses reported in the subsequent chapters of this book. At this point, however,
the primary task is to draw a clear dividing line between democracies and
authoritarian regimes, and this poses a serious methodological problem of
transforming the continuous measure into a dichotomous one (Bogaards,
2012). The theoretical limits of the values of the electoral democracy index
are from zero to one. In a recent study specifically dealing with the issue of
how to understand these values in binary terms (Lührmann et al., 2018),
the value of 0.5 is identified as a satisfactory cut-off point, but it is also rec-
ognized that in some cases, such a threshold can produce misleading results
as it is “admittedly arbitrary.” This resonates with the earlier study of Boix
et al. (2013, pp. 1536–1537) who mentioned that the 0.5 cut-off point, while
popular because of its intuitive appeal, actually lacks any theoretical grounds.
In some other studies, the threshold value of 0.4 for related V-Dem indi-
ces is used (Wang et al., 2019; Anckar, 2020). This approach is consistent with
the results of an in-depth analysis of Kasuya and Mori (2019). Proceeding
from the theoretically informed assumption that the preferred threshold
b4407_Ch02.indd 16 1/6/2022 9:14:50 AM

should achieve the highest consistency with the binary measures’ coding,
they employ a rather sophisticated methodological toolkit to establish that
for V-Dem, such a threshold can be set at the 0.42 level. My qualitative
investigation of the available cases that are relevant for the goals of this study
has demonstrated that in some instances, the distance between the values of
0.40 and 0.42, however small in numerical terms, can be highly consequential
in the process of dichotomization, and that the 0.4 cut-off point is better
suited for distinguishing between democracies and authoritarian regimes.
Keeping in mind that small numerical differences among values generated
by statistical analyses can result from data noise (Wolff et al., 2011), and that
even if performed in a methodologically sophisticated way, transforming
continuous measures into dichotomous ones invariably involves a degree
of arbitrariness, I have chosen to use the 0.4 cut-off point.
For the whole period of 1945–2019 covered in this study, the overall
number of country-year observations that satisfy this numerical criterion is
5762. The values of electoral democracy index for these observations vary
from 0.008 to 0.4. Of course, the number of elections held in these years
is much smaller, and the number of those elections that can be viewed as
manifestations of authoritarian party systems is smaller still. Yet, in order
to identify the set of eligible cases, it is necessary to define party systems in
both substantive and operational terms. This is particularly important for
performing the crucial task of case selection for empirical analysis.
Party Systems
Political parties are commonly defined as groups of candidates who compete
in elections (Schlesinger, 1994; Aldrich, 1995). Epstein (1980) supplements
this definition with an important additional component by specifying that
for a group of candidates to be qualified as a party, they have to run in
elections under a joint label. This definitional element refers to the fact that
the primary functions of political parties are double fold: on the one hand,
they provide coordination among politicians who join them in order to
b4407_Ch02.indd 17 1/6/2022 9:14:50 AM

6"×9"
compete in elections; on the other hand, they provide a wider coordination

that involves both politicians and voters, for which purpose party labels
are indispensable (Cox, 1997). Using the terminology of a different but
equally influential theoretical approach to political parties, they serve as a
link between the state and civil society (Lawson, 1976; Mair, 1994). Both
approaches emphasize, albeit from different perspectives, the role of parties
in state–society relations.
The party system is conventionally defined as “the system of interactions
resulting from inter-party competition” (Sartori, 1976, p. 44). This definition
is broad. It does not presume any properties of elections in which political
parties compete, and therefore does not restrict the notion of the party sys-
tem to democracies with their presumably free and fair elections. For a long
time, the broad definition of the party system remained widely recognized
but effectively ignored in political science because only democratic party
systems were considered as consequential and worth research effort. Albeit
not without notable exceptions (Janda, 1980), authoritarian party systems
did not receive much attention in the scholarly literature.
However, there are no solid theoretical reasons to deny that political
parties can compete in elections that are not free, fair, or, to put it in the terms
of Downs (1957, p. 25), “duly constituted.” It has been long recognized in
the literature that in authoritarian states, political parties and party systems
do play a role in channeling state–society relations (Neumann, 1956). Of
course, the roles of party systems in authoritarian and democratic politics
are different simply because unlike in democracies, autocracies do not allow
for alternation in power by means of elections. This obviously exerts a heavy
impact on the nature of inter-party competition.
But some of the substantive characteristics of competition do not wither
away under autocracy. In this connection, it is important to take into account
that competition is a complex concept embracing two components, contest
and cooperation (Franzmann, 2011). Under authoritarianism, the element
of contest is restricted but not entirely absent because access to some of the
government positions, albeit secondary, can be open to electoral contestation,
b4407_Ch02.indd 18 1/6/2022 9:14:50 AM

while the space for inter-party cooperation, as one would argue from the
perspective of cooptation, may be greater than in democratic conditions.
There is significant empirical evidence that limited contest and cooperation
can co-exist rather harmoniously in authoritarian party systems (Wong and
Or, 2020). The fact that cooperation is imposed upon political parties by the
dictatorial executive is certainly consequential. But even in democracies, the
role of the state in facilitating cooperative interactions among political parties
is quite pronounced, as reflected in the influential concept of the cartel party
(Katz and Mair, 1995). These considerations lead me to the conclusion that
the broad definition of the party system is applicable to the conditions of
authoritarianism, which implies a rather inclusive approach to case selection,
yet, at the same time delineates the scope of necessary exclusions.
First of all, the existence of elections does not necessarily imply the
existence of party systems or even political parties as such. Even among
democracies, there are several instances of electoral politics devoid of party
competition, which can be observed mostly in small island states (Anckar
and Anckar, 2000), even though there are notable historical precedents of
larger democratic polities without political parties, such as the 19th century
Orange Free State (Bryce, 2008, p. 266). Yet, such cases are rare, even though
the relative importance of independents in some democratic elections can
be substantial (Brancati, 2008b). The lack of explicit partisanship in author-
itarian elections is a widespread phenomenon.
Obviously, elections that are conducted on a non-partisan basis cannot
be considered as manifestations of party systems. Of course, this is not typical
of party regimes, even though it can be noted that one of the remaining com-
munist dictatorships, Cuba, conducts its elections on a formally non-partisan
basis (Roman, 1999). The lack of organized, legally permitted parties is quite
characteristic of monarchies, as illustrated by the cases of several Persian
Gulf states (Zaccara, 2013). Military regimes and personalist dictatorships
tend to be more tolerant toward political parties, but non-partisan elections
under these regimes are not uncommon. Thus, the non-partisanship cri-
terion serves as a reason for a large number of exclusions, including such
b4407_Ch02.indd 19 1/6/2022 9:14:50 AM

6"×9"
diverse cases as the Panchayat system in Nepal (Brown, 1996), the lasting
predominance of pro-regime independents in the electoral politics of Jordan
(Lust-Okar, 2009) and Belarus (Ash, 2015; Bedford, 2017), and the fractious
along ethnic lines, yet almost entirely lacking a party component, politics
of Afghanistan (Mobasher, 2019).
From the perspective of case selection, it is important to estab-
lish a numerical criterion that would allow for distinguishing between
party-structured and non-partisan elections. In this study, proceeding
from the common-sense assumption that non-partisanship prevails in the
conditions when no party representatives are needed for forming a major-
ity coalition in the assembly, I set the threshold at a rather high level of
50 percent. All elections in which party candidates gain less than a half of
seats are treated as non-partisan and thereby excluded.
Second, when selecting cases for inclusion into this analysis, I have to
choose among multiple layers of electoral politics, both vertical (national
versus sub-national) and horizontal (political executives versus legislatures).
My preference for national elections is only natural because this study is
focused on national party systems, although the unevenness in support of
political parties within nation states is considered as an important party
system parameter and thus accounted for in the further analysis.
The horizontal dimension presents a more difficult choice, but it has to
be taken into account that many autocracies do not hold direct chief executive
elections at all, while in some others, winning candidates prefer to run as
independents even if the assemblies are fully structured along party lines.
Thus, for the sake of data consistency, cases are selected exclusively on the
basis of national legislative elections. Only direct elections to lower or single
chambers of the assemblies have been taken into account. Elections to con-
stituent assemblies are included only on the condition that such assemblies,
upon completing their primary task of constitutional amendment, continued
to serve as regular legislatures.
Third, the theoretical notion of the party system, as outlined above, is
not applicable to those situations in which only one party is legally allowed
b4407_Ch02.indd 20 1/6/2022 9:14:50 AM

to participate in elections. This aspect was emphasized already by Sartori

(1976) who argued, in a rather self-explanatory way, that no system in the
true sense of the word can consist of only one component. Thus, I excluded
all elections in which inter-party competition was not legally permitted even
if it was possible for candidates to compete on an individual basis.
I extend this approach to those regimes where multiple parties are
allowed but do not compete in elections because the allocation of seats is
determined by pre-established quota systems. This used to be the case in a
majority of the communist regimes of the past. Of course, non-communist
parties were tolerated by such regimes on the condition of absolute political
loyalty, and their primary function was to broaden the “social bases” of the
authorities. In this capacity, such parties did play a role in the functioning
of the political system, at times receiving rather generous shares of seats in
regime-controlled legislatures, which was best exemplified by East Germany
(Allinson, 2000). Even today, China and North Korea formally have multi-
ple parties. Similar models of “multipartism” could be observed in several
non-communist party dictatorships, such as in Iraq and Syria under the rule
of the Arab Socialist Ba’ath Party (Perthes, 1993). A substantively very differ-
ent but typologically similar situation occurred in 1958–1974 in Colombia
where legislative seats, as well as other public offices, were divided in equal
proportions between the two main parties irrespective of election results
(Hartlyn, 1988). This analysis excludes all elections in which seat allocation
did not even formally depend on the expressed will of the voters.
Finally, the largest category of exclusions results from the fact that
party systems are recurrent, not situational, patterns of inter-party com-
petition. The recurrent nature of these patterns can be revealed only by
observing sequences of elections. Obviously, single elections do not suffice.
Sartori (1976) wrote about “atomized pluralism” in which parties do exist
and compete but do not interact in a patterned way. Following this line of
reasoning, Sanchez (2009, p. 487) introduces the concept of party non-systems,
defined as “party universes characterized by a fundamental absence of inter-
temporal continuity in the identity of the main parties.” In the contemporary
b4407_Ch02.indd 21 1/6/2022 9:14:50 AM

6"×9"
world, the salience of party non-systems greatly increased with the arrival of
many new democracies. As it will be shown in what follows, authoritarian
parties, and thereby, the patterns of interactions among them, also can be
extremely unstable.
Thus, the concept of a party system necessarily involves cross-
temporal continuity. From the angle of building operational criteria, the first
practical question is: how many elections are sufficient for a party system to
be observed? The answer to this question can arrive in no other form but
an arbitrarily set number. When identifying a criterion for viewing political
parties and their interactions as stable, many scholars (Rose and Mackie,
1988; Weber, 2011) follow Sartori (1976) who sets this number at three, even
though it is not unusual to establish more restrictive criteria such as four or
five elections (Bogaards, 2004; De Jager and du Toit, 2013). When choosing
among these approaches, I side with Siaroff (2019, p. 74) who argues that a
lesser duration should suffice. Thus, the minimum duration of a party system
is set as three consecutive elections.
The notion of sequence implies the lack of interruptions, which makes
it necessary to identify the varieties of interruptions that make this notion
empirically inapplicable. Following the set of criteria established above,
I considered as interrupted those sequences in which party-structured elec-
tions were interspersed with elections contested mostly by independents,
or by elections held on a single-party/national front bases, as described
above. Exceptions were occasionally made for those elections in which their
effectively single-party nature stemmed from the opposition parties’ choice
to withdraw from the races voluntarily by boycotting them, provided that
they remained legally eligible for participation.
Since this study deals with authoritarian party systems, one of the cri-
teria for exclusion is the lack of interruptions in the authoritarian character
of elections. Throughout the period under observation, many important
countries, such as Pakistan, continuously alternated between democracy and
authoritarianism. Some of the Pakistani political parties survived through
these repeated alternations, which explains a significant degree of continuity
b4407_Ch02.indd 22 1/6/2022 9:14:50 AM

in their development (Mufti et al., 2020). But the party system as such can-
not be characterized as continuously authoritarian because the patterns
of inter-party interactions were strongly influenced by the conditions that
existed during the democratic episodes. In operational terms, any increase
of the index of electoral democracy above the 0.4 threshold was considered
as an interruption.
To be realistically considered as a recurrent pattern of interactions, a
party system should involve a set of stable political parties. If no such parties
exist, so that every new election differs from the previous one in terms of
the composition of the field of political alternatives, then no party system
can be observed. The related criterion for exclusion draws on the notion
of extra-system volatility advanced by Mainwaring et al. (2010). Extra-
system volatility is defined as the share of votes or seats gained in the given
elections by new political parties, that is, by those parties that did not gain
seats in the previous elections. The number of observations for calculating
extra-system volatility is always (n − 1), where n is the number of elections
in the sequence. In this study, I excluded from the category of party systems
all those sequences in which average extra-system volatility for the whole
sequence exceeded 50 percent. This criterion is admittedly generous, as a
result of which it accounts for a small number of exclusions. Consistent with
the overall approach to electoral statistics that will be explained elsewhere
in this book, all calculations of extra-system volatility were performed on
the basis of seat shares rather than vote shares.
The criteria outlined above were instrumental in identifying the party
systems dealt with in this study, numbering 57 and comprising 294 elections
held in 55 countries from 1945–2019. It has to be emphasized that these
57 units are not a selection but rather the general population of cases satisfying
the criteria outlined above. As a rule, party systems are identified and referred
to as individual country cases. El Salvador stands as an exception to this
rule because my in-depth examination of this case has demonstrated that the
criteria for inclusion, especially with respect to extra-system volatility, are
best satisfied by separating from each other two of the country’s authoritarian
b4407_Ch02.indd 23 1/6/2022 9:14:50 AM

6"×9"
Table 2.1. Authoritarian party systems of the world, 1945–2019.

Most Average index
Number of Earliest recent of electoral
Party system elections election election democracy
Algeria 5 1997 2017 0.341
Angola 4 1992 2017 0.229
Armenia 4 2003 2017 0.372
Azerbaijan 5 1995 2015 0.228
Bolivia 5 1956 1964 0.263
Brazil 5 1966 1982 0.193
Burundi 3 2005 2015 0.283
Cambodia 5 1998 2018 0.319
Cameroon 5 1992 2013 0.319
Chad 3 1997 2011 0.283
Colombia 4 1945 1951 0.221
Republic of Congo 4 2002 2017 0.241
Cote d’Ivoire 3 1990 2000 0.262
Djibouti 6 1992 2017 0.253
Dominican Republic 3 1966 1974 0.257
Egypt 9 1976 2010 0.215
El Salvador (1961–1978) 9 1961 1978 0.182
El Salvador (1982–1991) 4 1982 1991 0.193
Equatorial Guinea 6 1993 2017 0.181
Ethiopia 5 1995 2015 0.251
Gabon 5 1990 2011 0.329
Gambia 4 1997 2012 0.266
Guatemala 4 1966 1978 0.164
Guyana 3 1973 1985 0.280
Indonesia 6 1971 1997 0.201
Iran 5 1956 1971 0.087
Kazakhstan 5 1999 2016 0.250
Madagascar 3 1960 1970 0.222
Malaysia 12 1964 2013 0.278
Mauritania 3 1992 2001 0.272
Mexico 15 1929 1988 0.260
Morocco 7 1984 2016 0.249
b4407_Ch02.indd 24 1/6/2022 9:14:50 AM

Table 2.1. (Continued)
Most Average index
Number of Earliest recent of electoral
Party system elections election election democracy
Nepal 3 1991 1999 0.365
Nicaragua 5 1950 1974 0.166
Paraguay 6 1963 1988 0.166
Philippines 4 1946 1957 0.327
Portugal 7 1949 1973 0.131
Russia 4 2003 2016 0.313
Rwanda 4 2003 2018 0.217
Serbia 3 1992 1997 0.265
Seychelles 3 1993 2002 0.360
Sierra Leone 3 1967 1977 0.247
Singapore 12 1968 2015 0.360
South Africa 11 1948 1989 0.186
South Korea 5 1963 1978 0.250
South Vietnam 3 1956 1963 0.299
Sudan 3 2000 2015 0.204
Taiwan 7 1972 1992 0.171
Tajikistan 4 2000 2015 0.222
Thailand 6 1975 1988 0.287
Togo 4 1994 2007 0.318
Tunisia 7 1981 2009 0.192
Uganda 3 2006 2016 0.356
Uzbekistan 4 2004 2019 0.180
Yemen 3 1993 2003 0.314
Zimbabwe (Rhodesia) 3 1970 1977 0.198
Zimbabwe 8 1985 2018 0.278
party systems, chronologically defined as 1961–1978 and 1982–1991, which

is how these units of analysis are referred to in the following. The case of
Zimbabwe is also distinct, and for a rather apparent reason, it had to be split
into the racial segregation-era party system of 1970–1979, further referred to
b4407_Ch02.indd 25 1/6/2022 9:14:50 AM

6"×9"
as Zimbabwe (Rhodesia), and the post-segregation system, f urther referred

to as Zimbabwe.
Table 2.1 lists the party systems in the alphabetical order. In addition, the
table reports the numbers of elections in which these systems were manifest
and the years of the earliest and latest elections throughout the periods of
their existence. The average index of electoral democracy is calculated on
the basis of its V-Dem values for the years of elections.
b4407_Ch02.indd 26 1/6/2022 9:14:50 AM

Chapter 3
The World of Authoritarian

Party Systems
T
he purpose of this mostly descriptive chapter is to provide essential
factual information about the universe of authoritarian party sys-
tems. The emerging scholarship on electoral authoritarianism has
been focused either on the relatively recent cases, such as the states of the
former Soviet Union, or on particularly long-standing electoral autocracies,
such as Mexico and Malaysia, even though the emphasis was again placed
on the relatively recent periods of their existence. A majority of the author-
itarian party systems of the past did not receive much scholarly attention
in the times when they were in place, mostly because political scientists
rendered little or no importance to them, and some of these systems have
now become completely obliterated. Meanwhile, authoritarian party systems
are neither new nor rare. Table 2.1 clearly demonstrates that throughout the
period of 1945–2019, they made their appearance in more than a quarter
of the world’s countries.
The geographical spread of authoritarian party systems throughout
the period was uneven, and they were unevenly distributed over time.
This is shown in Table 3.1 that presents the distribution of authoritarian
party systems by the regions of the world and by two chronological periods
established in accordance with the waves of democratization, as defined
by Huntington (1993). At this point, Huntington’s concept is employed
simply as a tool of periodization, even though the substantive link between
the waves of democratization and the spread of electoral authoritarianism
has been noted in the literature (Magaloni and Kricheli, 2010) and will be
27
b4407_Ch03.indd 27 1/6/2022 9:15:02 AM

6"×9"
Table 3.1. The geographical and chronological distribution of authoritarian

party systems.
Region 1945–1973 1974–2019 Total
West Europe 1 — 1
America 10 1 11
East Asia 7 2 9
Middle East and South Asia 1 2 3
Africa 4 22 26
Post-Communist 0 7 7
Total 23 34 57
discussed further in this chapter. In particular, the table demonstrates that

after the third wave of democratization, authoritarian party systems found
Africa and the post-Communist world as their primary location, but in the
course of the previous period, they were mostly located in America and
East Asia. The only West European case, Portugal, not only belonged to the
early period but also, as a result of its termination in 1974, defined its upper
chronological limit. This explains both the order of the regions in the table
and the order of the descriptive sections of this chapter.
Figure 3.1 clarifies the information presented in the table by showing
that the spread of authoritarian party systems across time accelerated during
two periods, in 1965 through 1977 and in 1989 through 2007. This suggests
that while the rise of the phenomenon coincided with the attenuation of the
second wave of democratization rather than with its ascent time, the major
impetus for the emergence of authoritarian party systems during the third
wave was related to, if not stemmed from, a certainly democratizing event
such as the collapse of global communism.
It has to be noted that the visualization provided in Figure 3.1, while
indicative in many respects, can be misleading in two others. First, the figure
underscores the spread of authoritarian party systems in 2008–2019. This
may evoke a misleading impression that the phenomenon starts to perish.
But, while it is true that some of the authoritarian party systems did recently
cease to exist, several others are in the making, which will be discussed at
b4407_Ch03.indd 28 1/6/2022 9:15:02 AM

The World of Authoritarian Party Systems 29
35
30
25
20
15
10
0
1945 1955 1965 1975 1985 1995 2005 2015
Figure 3.1. The historical dynamics of authoritarian party systems:

Numbers of country cases in existence by year, 1945–2019.
some length in Chapter 8 of this book. The emerging authoritarian party

systems remain unregistered in Figure 3.1 simply because they did not have
enough time to survive through three consequent elections. But the prospects
for these systems’ long-term survival are not necessarily negligible.
Second, it may appear from the table that since authoritarian party
systems were so rare in 1945, the pre-1945 cases have to be episodic. But
they are not. In fact, the historical antecedents of authoritarian party systems
are not very numerous, but no less than fifteen cases can be registered with
confidence by using the same criteria as laid out in the previous chapter of
this book, and some of these cases are important for our understanding
of the current trends. This explains why this chapter, while presenting its
descriptive content mostly by region, includes two chronologically defined
opening sections. One of them briefly overviews the pre-World War II
antecedents, and the other includes information about those authoritarian
party systems that were in place already in 1945–1949. Some of these systems
bear resemblance to the pre-1945 phenomena.
b4407_Ch03.indd 29 1/6/2022 9:15:02 AM

6"×9"
Historical Antecedents
The historical antecedents of authoritarian party systems can be divided into
several categories, and not all of them present primary interest for this study.
Technically, for instance, the party system of the United States existed in an
authoritarian context from 1825, when it consolidated in a bipartite form
after the “era of good feelings” (Stonecash and Brewer, 2009), to 1897, when
the United States passed the 0.4 threshold on the V-Dem scale of electoral
democracy. The limitations on suffrage and political competition in the
US were strong enough to justify the values of the index, averaging 0.351
throughout the period, but the impact of these factors upon party system
development can be better traced by focusing on different country cases.
Two categories of the authoritarian party systems of the past, those of
the 19th–early 20th century empires and those of the European monarchies
of the same period, with their greater degrees of political freedom but very
limited suffrage, are of a more immediate interest for this study. It has to be
noticed that these two categories are cross-cutting, but they can be analyt-
ically distinguished because of the differences on theoretically important
parameters.
The party systems of the 19th–early 20th century empires are best
exemplified by the case of Germany. In 1871, when the German Empire
(Kaiserreich) was established, a system of universal male suffrage was intro-
duced and implemented consistently throughout the whole history of the
empire (Senigaglia, 2020). The political role of the parliament was however
limited because its legislative powers were curtailed by the non-elected
Federal Council, and even to a greater extent, because there was no govern-
ment responsibility to the parliament. The political executive was dominated
by the Emperor and by the Prime Minister (Reichskanzler) appointed by him
without parliamentary consent (Orlow, 2018). These conditions, while thor-
oughly authoritarian, with the average index of electoral democracy equaling
0.238 for 1871–1918, did not create much incentive for forming a dominant
party in order to control the elected parliament on behalf of the executive.
b4407_Ch03.indd 30 1/6/2022 9:15:02 AM

At the same time, it was possible to endow the assembly with the functions
of representing different societal constituencies (Anderson, 2000).
Indeed, the party system of Imperial Germany featured a plethora of
political parties representing different segments of the political spectrum,
from a number of pro-government conservative and liberal parties to the
rather powerful but fragmented opposition, including the clerical German
Center Party and the Social Democratic Party of Germany that rose to
political prominence in the end of the 19th century. None of these parties
ever obtained a majority of seats in the parliament. As put by Sperber (2003,
p. 364), “the executive did not need a regular parliamentary majority to pass
legislation, but could form ad hoc majorities from bill to bill.” Similarly frac-
tious party systems had developed in Russia after the partial liberalization
of 1905 that brought to life a national assembly with very limited legislative
powers, the Duma (Hosking, 1973), and in the early 20th century Austro-
Hungary (Cohen, 2007).
An alternative route of party system formation in the authoritarian con-
text is represented by those West European monarchies where the gradual
introduction of parliamentary government occurred in parallel with the
incremental extension of suffrage (von Beyme, 2000). This category includes,
among others, the cases of Belgium, Denmark, and the United Kingdom.
The processes of early party system development in such polities have been
thoroughly examined in a number of richly empirical studies (Bartolini and
Mair, 1990; Caramani, 2004; Miller, 2015a). The theoretical logic, as revealed
in these studies, can be briefly explicated as follows.
On the one hand, the early introduction of the elements of parliamentary
government would have made fractious party systems dysfunctional. Political
parties emerged as rather inclusive coalitions that could compete not only for
parliamentary seats but also, effectively, for a large share of executive power.
On the other hand, limited suffrage constrained the spectrum of societal
interests represented in the party system to a small number of propertied
classes, often including the landed gentry and urban middle class. This led
to the emergence of two-party systems in the United Kingdom, with the
b4407_Ch03.indd 31 1/6/2022 9:15:02 AM

6"×9"
Conservatives and the Liberals, and in Denmark, with major parties labeled
simply the Left and the Right. In Belgium, a very salient confessional cleavage
generated a recurrent pattern of competition between the Catholic Party
and the Liberal Party.
In some cases, progressive democratization, in combination with the
extension of suffrage that led to the emergence of socialist/workers’ parties,
broke this pattern and led to the emergence of democratic multipartism,
which was typical of the majority of West European countries. In other cases,
however, bipartite competition survived well into the post-authoritarian
periods of political development, albeit with a different composition of
leading political actors, which is of course best exemplified by the United
Kingdom. It is interesting to note that the only pre-1945 authoritarian
party system in Asia, the party system of Japan after the Meiji restoration,
displayed a sequence of two developmental patterns. In its original form,
the Japanese party system was rather fractious, which reflected the effective
predominance of the executive in the political system. But as the effective
powers of the Japanese Diet gradually increased in the 1920s, a new pattern
of bipartism started to emerge (Fukui, 1988).
The inter-war period yielded few authoritarian party systems. While
some of the European countries democratized, the right-wing dictatorships
that became widespread throughout the period, and especially in the 1930s,
normally excluded all political opposition from running in elections, as
exemplified not only by the paradigmatic cases of Fascist Italy and Nazi
Germany but also by pre-Anschluss Austria, Poland, and the Baltic countries
in the 1930s. It is however notable that in some cases, such as in Hungary
under the dictatorship of Miklós Horthy (Janos, 1970), limited party compe-
tition was tolerated. In fact, the structural characteristics of the Horthy-era
party system of Hungary, with its outright dominance of the pro-Horthy
National Unity Party and visible but fractious opposition, are remarkably
similar to the specific type of authoritarian party systems that became most
widespread in the post-war period, particularly in Africa and in the former
Soviet Union.
b4407_Ch03.indd 32 1/6/2022 9:15:02 AM

Early Contemporary Cases

Only two right-wing dictatorships established in the 1930s survived well
into the post-war period. One of them, Spain, remained a single-party
state, even though the role of the sole legal and formally ruling party, the
Traditionalist Spanish Phalanx and (that) of the Councils of the National
Syndicalist Offensive, gradually declined over time (Payne, 1987). The
second case is Portugal under the long-standing dictatorship of António
Salazar. From 1932 to 1945, the Salazar regime was a one-party state in
which only the pro-government organization, labeled the National Union,
was permitted. In 1945 and thereafter, opposition groups were allowed to
run their district-level lists in national legislative elections (Raby, 1989),
which qualifies Portugal as a case of electoral authoritarianism. However,
opposition lists were invariably nominated in a small minority of electoral
districts, and some of them were disqualified from running by the author-
ities. In fact, not a single seat in the national legislature was gained by the
opposition in the regime’s lifetime. Even in these conditions, some of the
elections produced results that revealed some degrees of discontent in
the electorate (Martins, 1969, pp. 253–254).
Several early authoritarian party systems can be observed in Latin
America. In fact, Latin America is home to a number of the oldest entities
calling themselves political parties, some of them tracing their origins to the
first half of the 19th century (Alexander, 1982). But for many of these entities,
electoral arenas were secondary at best because their power struggles were
waged mostly by means of military coups, uprisings, and intra-elite deals.
Such patterns of inter-party competition cannot be viewed as party systems.
In the course of time, however, a number of countries, such as Colombia
and Nicaragua, did develop authoritarian party systems consisting of their
historical parties.
After a long history of acute fights for power between the Conservative
and Liberal parties of Colombia (Deas, 1996), and upon the completion of a
period of the effective political monopoly of the Liberals, the country entered
b4407_Ch03.indd 33 1/6/2022 9:15:02 AM

6"×9"
a phase of electoral authoritarianism when both historical parties competed

in the electoral arena, which for some time continued even in the conditions
of the civil war, La Violencia, that broke out in 1948 and eventually led to a
military coup. In Nicaragua, one of the two historical parties, known as the
Nationalist Liberal Party during the post-war period, was effectively appro-
priated by the Somoza family that ruled the country in 1936 through 1979,
while the second historical party, the Conservatives, performed the role of
the officially permitted and almost invariably loyal opposition (Hoyt, 2004).
It was only in 1967 that a less harmless opposition coalition, the National
Opposition Union, was allowed to participate in elections.
When characterizing the early political parties of Latin America, it is
common to mention the domination of oligarchic groups that practically
proscribed the masses from political participation even under the conditions
of a legally inclusive franchise (Martz, 1964). A similar situation occurred
in the Philippines where, briefly after the end of the Japanese occupation,
a split within the historical and once dominant Nationalist (Nacionalista)
Party, founded in 1907, generated the Liberal Party. This opened the way
for a largely bipartite pattern of electoral competition that eventually led to
a temporary democratization in the second half of the 1950s. Both parties
were non-programmatic elite groups whose electoral success was largely
predetermined by control exercised by the landed/propertied classes over
the political behavior of the economically underprivileged (Hutchcroft and
Rocamora, 2003), which makes the Philippine experience broadly similar
not only to the Latin American cases but also to the early patterns of party
politics in some of the West European countries.
A similarly dominant role of economic elites, commonly referred to
as Caciquismo (Knight, 2005), can be observed in Mexico that hosted the
most durable and stable authoritarian party system of the post-war period.
Indeed, the case of Mexico is paradigmatic in contemporary research on
electoral authoritarianism (Magaloni, 2006; Greene, 2007). The ruling party
of the Mexican authoritarian regime, known for the most part of its history
as the Institutional Revolutionary Party, was formed in the late 1920s as a
b4407_Ch03.indd 34 1/6/2022 9:15:02 AM

coalition of local elites that emerged from the sweeping revolution of the early
20th century. As soon as in the 1930s, several unstable opposition parties
emerged. By 1946, however, the set of these parties stabilized.
For a long time, most prominent roles were played by the National
Action Party, representing the right segment of the political spectrum, and
the Mexican Communist Party or its frontline organizations representing
the left-wing. Several other parties emerged and disappeared throughout the
period of the regime’s existence, but the electoral niches of the main opposi-
tion parties remained small, albeit never negligible, up to the 1980s. It was
not until 2000 that the ruling party, after prolonged political struggles that
were waged for a large part outside of the electoral arena, conceded defeat
to the presidential candidate of the National Action Party, which led to the
end of electoral authoritarianism in Mexico (Chand, 2001).
The authoritarian party system that existed in South Africa throughout
the apartheid era of its history stands on its own. The reason for its peculiarity
is, of course, the apartheid itself. The system of racial segregation not only
restricted suffrage to a minority of the population but also created a stable
political cleavage between the Afrikaans-speaking population, who tended to
support the apartheid, and the English-speakers who tended to take a more
liberal stance. In these conditions, the National Party that drew its support
mostly from the Afrikaner population continuously received legislative
majorities. Liberal parties, drawing their support from the English-speaking
white minority, operated without facing serious political constraints, but
without any prospects for breaking the domination of the National Party
(Mark and Trapido, 2014). The collapse of apartheid naturally resulted in
the collapse of this system.

the Western Hemisphere
For the most part, the Latin American authoritarian party systems existed
under the auspices of military dictatorships that constituted a widespread
b4407_Ch03.indd 35 1/6/2022 9:15:02 AM

6"×9"
variety of non-democratic rule in the region. Many of these dictatorships

did not need political parties because they were short-lived and did not
conduct elections for this reason alone, even though some of them, such as
the left-wing regime of Juan Velasco Alvarado in Peru, avoided party poli-
tics for programmatic reasons as they sought a different model of political
mobilization (Stephens, 1983). Others, including the dictatorship of Augusto
Pinochet in Chile, claimed that any political opening had to be preceded by
the ultimate uprooting of communism. But there were exceptions.
One of them, Paraguay, stands close enough to some of the cases
reviewed above. Similarly to the Somoza family in Nicaragua, the long-
standing dictator of Paraguay, Alfredo Stroessner, utilized a historical party,
the National Republican Association — Colorado Party, as his electoral
vehicle. In the course of nearly a decade after Stroessner’s coming to power,
Paraguay was effectively a single-party state. Starting with 1963, however,
several other parties were allowed to run in elections. They jointly received
a guaranteed third share of legislative seats thereafter in a system that lasted
up to 1989 when Stroessner was ousted from power by a military coup
(Nickson, 1989).
In several other countries, the military were either unable to utilize
historical parties or simply did not have such an option because of these
parties’ absence or disappearance. In these conditions, new political par-
ties created and directly controlled by the military were placed at the core
of authoritarian party systems. This was the case in El Salvador with its
National Conciliation Party that dominated the political arena from 1961
to 1976, and in Dominican Republic with its Reformist Party that served
as an electoral vehicle of the successor to the Trujillo family dictatorship,
Joaquín Balaguer. In both countries, the electoral participation of moderate
opposition parties was allowed and occasionally brought them significant
shares of legislative seats even despite the fact that these parties were sub-
jected to systematic “diversion, subversion and repression,” as noted by one
observer of the Dominican Republic politics in the 1970s (Kryzanek, 1977,
p. 83). It is notable that the second authoritarian party system of El Salvador,
b4407_Ch03.indd 36 1/6/2022 9:15:02 AM

generated by the military coup of 1978, was based on competition between

the National Conciliation Party and the Christian Democratic Party that
supported the new coup (Karl, 1989).
An even more interesting case of a thoroughly authoritarian yet to an
extent competitive system is presented by Guatemala where a succession
of military dictatorships created several parties that formed the core of the
country’s party system in 1966–1978. The Institutional Democratic Party,
modeled after the then-ruling party of Mexico, was created by Enrique
Peralta Azurdia after he had seized power in the coup of 1963. When the
party entered the electoral arena in 1966, it had to confront two parties
inherited from the previous periods of the country’s turbulent political
development, the center-left Revolutionary Party and the National Liberation
Movement created by the previous military dictator. While allowing for mul-
tiparty elections and even alternation in power among the permitted parties,
the Guatemalan military relied on massive violence and cruel repression
against the banned opposition parties in order to retain effective control
over the state (Schirmer, 1998).
The most far-reaching and well-described in the literature (Hagopian,
1996; Mainwaring, 1999) attempt of creating a new party system by the ruling
military had been undertaken in Brazil after the coup of 1964. As originally
designed, the system was centered around the pro-regime National Renewal
Alliance, while those parties that were pivotal in the democratic politics
of Brazil in the 1950s and early 1960s were dissolved and then artificially
merged into the only opposition party permitted by the regime, the Brazilian
Democratic Movement. Highly repressive in the early years of its existence,
the Brazilian military regime did not rely on massive electoral fraud (Codato,
2006). As the popularity of the regime’s “law and order” agenda gradually
vanished, the opposition party’s shares of the vote in the national legislative
elections of the 1970s became dangerously close to those of the dominant
party. In an attempt to prevent an outright loss in bipartite competition,
the regime legalized several opposition parties, both historical and new. By
the middle of the 1980s, a split in the ruling party finalized the demise of
b4407_Ch03.indd 37 1/6/2022 9:15:02 AM

6"×9"
the dictatorship that was already shaken by massive industrial unrest and
electoral decay at the sub-national level.
Two other countries in the Western Hemisphere represent rather idio-
syncratic patterns of authoritarian party system development. In Bolivia, the
National Revolution of 1952 brought to power the Revolutionary Nationalist
Movement. After a brief period of single-party rule (Malloy and Thorn,
1971), elections with the participation of several other parties were rein-
stituted, but the political predominance of the Revolutionary Nationalist
Movement remained unbroken until it had been terminated by a military
coup in 1964.
In Guyana, the authoritarian party system emerged as a result of acute
power struggles between two parties representing the country’s ethnic
communities, the People’s National Congress supported primarily by the
Afro-Guyanese community and the People’s Progressive Party that mostly
drew support from the Indo-Guyanese. Several years after coming to power
in 1964, the People’s National Congress started to display increasingly
authoritarian tendencies, using election fraud, control of the media, and
other less than democratic means in order to prevent the major opposition
party from winning at the polls (Taylor, 2019). But the shares of legislative
seats controlled by the opposition remained quite significant. In the begin-
ning of the 1990s, the outright political domination of the People’s National
Congress came to an end as a result of free elections.
The Authoritarian Party Systems of Asia

A distinct group of authoritarian party systems emerged in three East Asian
countries that confronted existential threats in the form of alternative com-
munist regimes competing for legitimacy: South Vietnam, South Korea, and
Taiwan. To a great extent, these countries’ willingness to conduct multiparty
elections stemmed from their official commitment to democracy as an ideo-
logical alternative to communism. Yet, in the 1950s through 1970s, they fell
short of converting this commitment into full-fledged democratic reforms.
b4407_Ch03.indd 38 1/6/2022 9:15:02 AM

This is particularly true of the highly repressive regime of Ngo Dinh

Diem in South Vietnam where the electoral scene of 1956–1963, while
dominated by the pro-regime National Revolutionary Movement, also
included several parties with less than credible opposition stances, such
as the one headed by the dictator’s brother (Scigliano, 1960). In a similar
vein, the long-standing dictator of South Korea, Park Chung Hee, created
the Democratic Republican Party primarily as a tool of his control over
the legislature. In contrast to South Vietnam, South Korean elections were
not entirely fraudulent; nor were these elections completely purged of the
democratic opposition parties (Heo and Stockton, 2005). One of them even
outperformed the Democratic Republican Party in terms of voter support in
the 1978 elections, even though the ruling party still won a majority of seats
with the help of pro-regime independents. In both countries, authoritarian
party systems were terminated by military coups.
The ruling party of Taiwan, normally referred to as the Kuomintang
and sometimes as the Nationalist Party of China, was imported to the island
after losing its fight against the Communists in mainland China. For a long
time, Taiwan did not have political institutions of its own. Starting with
the early 1970s, however, the number of representatives directly elected to
the legislature gradually increased, and they were elected on a multiparty
basis, which created an authoritarian party system still dominated by the
Kuomintang. It was only in 1989 that, for the first time, an opposition party
gained a sizeable minority of seats (Tien, 1996).
The party systems of Malaysia and Singapore took their roots in
the patterns of party politics that emerged in both countries during the
British colonial rule. Already in the 1950s, the United Malays National
Organization achieved political prominence in a majority of traditional
monarchies that were to form contemporary Malaysia. Later, this party,
drawing its support mostly from the ethnic Malay population, was joined
by several smaller ethnicity-based and/or territorial parties to form the
National Front (Weiss, 2014). The Democratic Action Party formed the
core of the opposition by appealing to different ethnic groups, primarily
b4407_Ch03.indd 39 1/6/2022 9:15:03 AM

6"×9"
the Chinese, but it came distant second in all elections held in 1964
through 2013.
It is interesting to note that the Democratic Action Party was an offspring
of the People’s Action Party that, after being banned in Malaysia, carried on as
the ruling party of Singapore after its independence in 1965 (Tan, 2014). For
a long time, the regimes of Malaysia and Singapore placed severe restrictions
on the activities of the opposition, which explains the authoritarian character
of their party systems. A recent unexpected turn in the political trajectory of
Malaysia, resulting from an effective split in the ruling party, led to a partial
and possibly unstable democratization, while the highly authoritarian traits
of the Singaporean party system remain largely in place, even though not
without some faint signs of liberalization (Ortmann, 2011).
Indonesia under the long-standing dictatorship of Suharto presents
one of the most important historical instances of electoral authoritarianism.
Upon coming to power in the military “counter-coup” of 1965, Suharto
assumed presidency in 1967 and, starting in 1971, authorized holding regular
elections in the country. Several parties inherited from the previous period
of Indonesia’s political development participated in the earliest of these
elections, but starting from 1977, only three parties were allowed to run:
the pro-regime formation largely controlled by the military, Golkar, which
is commonly translated as the Organization of Functional Groups; and two
parties that were seeking to represent the Islamic and secular segments of
opposition-minded voters, called the United Development Party and the
Indonesian Democratic Party, respectively (Crouch, 1979). This composition
of Indonesia’s party system remained unchanged until 1998 when, amidst
an economic crisis, Suharto was deposed in a military coup.
Ali Abdullah Saleh came to power in North Yemen in 1977 as a new
leader of the military regime that existed in the country since the 1960s. In
1982, Saleh launched a new party, the General People’s Congress, enjoy-
ing the status of the only legal political organization. At the same time, a
communist-type single-party regime of the Yemeni Socialist Party existed in
South Yemen. After the unification of the two states in 1990, Saleh remained
b4407_Ch03.indd 40 1/6/2022 9:15:03 AM

in power but allowed for holding multiparty elections with the participa-
tion of the former ruling parties, primarily for the purpose of determining
the quotas of ministers from each party in the government (Detalle and
Hiltermann, 1993). A party of moderate Islamists, the Yemeni Congregation
for Reform, was also allowed to run and eventually replaced the Socialists
as the second-largest party (Durac, 2011). The “Arab spring” revolution of
2011 brought this party system to its end.
The role of the military in politics was also quite visible in the develop-
ment of the authoritarian party system of Thailand. Since at least 1932, the
political system of the country, formally a parliamentary monarchy with a
significant constitutional role of the crowned head of state, was based on
a fusion between the interests of the royal court, propertied elites, and the
corporate interests of the military (Chambers and Waitoolkiat, 2016). For
a long time, Thailand’s party scene was dominated by pro-regime indepen-
dents and displayed extreme volatility. After the massive political unrest of
1973 and a resulting partial liberalization, the pattern of electoral compe-
tition stabilized in the form of a fractious system that included a plethora
of parties. Most of them were connected, in different ways and with their
private interests on the agenda, with the royal court, the military, and local
economic elites (Bunbongkarn, 1987; Laothamatas, 1988). In 1991, tensions
between the dominant military faction and the civilian government led to a
coup. In the ensuing period, characterized by frequent alternation between
democratic and authoritarian phases of political development, no stable party
system could be registered in Thailand, even though some of the parties that
emerged in the 1970s remained in place.
Two other Asian monarchies of the post-war period experimented with
authoritarian party systems. In Iran, the autocratic regime of Mohammad
Reza Pahlavi invested some effort into creating a system with two legally
permitted parties: the pro-regime organization known, during the most
recent part of its history, as the New Iran Party, and the People’s Party of
loyalist liberals. Created by a royal decree, this system was decreed into
non-existence and replaced by a single-party system several years before the
b4407_Ch03.indd 41 1/6/2022 9:15:03 AM

6"×9"
Islamic revolution in Iran (Azimi, 1997). Nepal is a sharply contrasting case.

As a result of the political unrest of 1990, King Birendra Bir Bikram Shah
was forced to initiate a constitutional transition to parliamentary monarchy
with rather limited powers of the crowned head of state (Khadka, 1993).
In the following years, three parties — the monarchists of the National
Democratic Party, the liberals of the Nepali Congress, and the Communist
Party of Nepal (Unified Marxist–Leninist) — competed for power in rel-
atively free elections, which opened a way towards democratization. In
the beginning of the 2000s, however, this party system collapsed due to a
combination of the Maoist insurgency and a temporary revitalization of
the monarchy (Hutt, 2006).
In Cambodia, the 1991 Paris agreements obliged the pro-Vietnamese
Communists to abandon the monopoly of power enjoyed by them from
1979 and to institute free elections. However, the ex-communist Cambodian
People’s Party used its vast political resources accumulated during the period
of single-party rule, as well as severe political restrictions, to sideline the two
other parties of the Paris Agreement, as a result of which the Cambodian
regime evolved into what is now one of the most salient examples of electoral
authoritarianism (Morgenbesser, 2019). This process culminated in the 2018
elections when, despite the participation of nineteen officially permitted
opposition parties, the Cambodian People’s Party was able to win all seats
in the national legislature.
The Authoritarian Party Systems of Africa

In the 1950s, major colonial states — primarily France and the United
Kingdom — granted internal autonomy to many of their African colonies,
which gave rise to political parties, including those that came to power after
independence. Very few party systems, democratic or authoritarian, were
produced by this trend. For the most part, the nascent African polities aban-
doned the practice of competitive elections in favor of single-party systems
or military regimes (Collier, 1978).
b4407_Ch03.indd 42 1/6/2022 9:15:03 AM

But there were exceptions. In post-independence Madagascar, the

regime of Philibert Tsiranana and his Social Democratic Party tolerated
legal opposition in the form of the leftist Party of the Independence
Congress of Madagascar. The support base of the opposition party was
narrow and urban. In the rural areas of the predominantly agricultural
country, the ruling party neutralized all potential opposition through a
combination of coercion and cooptation, which allowed for taking its elec-
toral success for granted (Marcus and Ratsimbaharison, 2005). This party
system collapsed as a result of a military coup. In Sierra Leone, a prolonged
struggle between the All-People’s Congress and Sierra Leone People’s
Party was settled by the decisive victory of the former in the relatively
free elections of 1967. In the course of the following decade, Sierra Leone
People’s Party was still allowed to participate in elections, even though the
increasingly authoritarian traits of the regime made alternation in power
quite unlikely. Then a single-party regime of the All-People’s Congress was
introduced (Luke, 1988).
In Zimbabwe, there were two authoritarian party systems naturally
divided by the transition to black majority rule in the end of the 1970s.
The ruling party of the early period, the Rhodesian Front, dominated the
political arena to the extent that it invariably gained all “European” seats
in the country’s legislature. This was of course related to its role as a pillar
of the segregation system that had been supported by a vast majority of
the white settlers (Lemon, 1978). After 1980, despite the Marxist–Leninist
ideological persuasion of the ruling Zimbabwe African National Union —
Patriotic Front of Robert Mugabe, multiparty elections were not abolished,
but initially, the role of the permitted opposition parties was modest at best.
In the end of the 1990s, the Movement for Democratic Change emerged as
a serious electoral challenge to the regime of Mugabe, but a combination
of centralized repression, election-time violence, and fraud (Alexander and
McGregor, 2013) made it possible for the regime to survive and even outlive
Mugabe. In 2018, he was removed from power by the ruling party with the
assistance of the military.
b4407_Ch03.indd 43 1/6/2022 9:15:03 AM

6"×9"
In the early 1980s, Arab-speaking North Africa emerged as a major

hub of electoral authoritarianism. The case of Egypt has received partic-
ularly extensive coverage in the literature (Kassem, 2004; Blaydes, 2010).
In the second half of the 1970s, the single-party system of the country
was replaced by a system in which the ruling party, later known as the
National Democratic Party, was supplemented by two opposition parties
of legally prescribed ideological stances, one to the right and the other to
the left of the presumably centrist ruling party. In the 1980s, the spectrum
of permitted political parties became wider, but a serious challenge to the
ruling party was occasionally posed only by the Muslim Brotherhood that
was able to run its candidates as independents (al-Awadi, 2009). Electoral
participation did not grant any significant representation to the Islamist
opposition but allowed for consolidating its support base to the extent
that it became a major force in the Egyptian revolution of 2011 and briefly
came to power in the post-revolution period. Broadly similar dynamics
could be observed in Tunisia where, after a period of single-party rule that
lasted up to 1981, the regime of the dominant Constitutional Democratic
Rally allowed for the electoral participation of several carefully selected,
presumably harmless opposition parties (Sadiki, 2002). In 2011, after a long
sequence of overwhelming electoral victories, the regime was overthrown
by a revolution.
The monarchical regime of Morocco formally instituted multiparty
elections as early as in 1963, but in the 1960s through 1970s, the electoral
arena of the country was dominated by pro-regime independents. It was only
in 1984 that this pattern was broken by the emergence of a party system,
comprising two overtly royalist organizations and a number of rather strong
opposition parties of national-liberal, moderate left, and later, moderate
Islamist persuasions. Even though not without some constraints, the regime
allowed these parties not only to compete in relatively free elections but also
to participate in the country’s government on the condition that they did not
question the royal monopoly on making decisions in major policy domains.
This was almost explicitly recognized in a deal made between King Hasan II
b4407_Ch03.indd 44 1/6/2022 9:15:03 AM

and then a major pro-democracy party, the Socialist Union of Popular Forces,
in the middle of the 1990s (Sater, 2009). It is common to consider Morocco
as a classical example of the efficient use of cooptation and manipulation
as primary tools of preserving power in an autocracy (Lust-Okar, 2004).
The global political upheaval of the late 1980s and early 1990s resulted
in the massive arrival of new authoritarian party systems in sub-Saharan
Africa. Many countries abandoned their previous single-party regimes of
different origins, mostly post-colonial or military, in favor of multiparty
elections. Some of these countries, such as Benin and Ghana, successfully
democratized. But many others evolved into electoral authoritarian regimes
(Manning, 2005). Several of them, including Angola, Cameroon, Djibouti,
and Equatorial Guinea, displayed very similar patterns of party system
development characterized by the outright domination of their ruling
parties inherited from the previous period of their political development.
These parties have simply preserved their domination in a political context
that has changed only in the sense that the opposition parties, while being
explicitly banned in the past, are now allowed to exist in the political niches
prescribed and carefully delineated by the authorities.
It has to be emphasized that while the four regimes listed above have
been continuously authoritarian up to the time of writing, several others
have not. Gabon and Togo initially followed the above-described pattern
but later experienced democratizations, albeit incomplete and reversible.
In both cases, these partial democratizations were prompted by change of
the executive leadership as a result of death of their long-standing dictators.
However, the hereditary successors of these dictators, while allowing for
some political opening immediately after the passing of their predecessors
(Nwosu, 2012), eventually consolidated their positions and reverted to the
old ways. As a result, authoritarian party politics in these countries resurged
after interruptions. One party system of this type, in Côte d’Ivoire, collapsed
in the beginning of the 2000s soon after a military coup followed by the
2002–2004 civil war (Daddieh, 2001). In fact, the Seychelles gives a unique
example of a former single party that, while retaining its power for a long
b4407_Ch03.indd 45 1/6/2022 9:15:03 AM

6"×9"
time after a partial democratization, did not preside over an autocratic turn
and ultimately lost power as a result of elections (Bhim, 2019).
In the Republic of Congo, the democratization of the early 1990s was
reversed as a result of the 1997 civil war (Bazenguissa-Ganga, 1999). The
political turbulence of the early 1990s led to a series of military coups and
civil wars that ultimately brought about several new authoritarian party
systems. In Algeria, the National Liberation Front, after a long period of
single-party rule, lost the first multiparty elections of 1991 to Islamists,
which led to a military coup and then to the civil war. When building the
authoritarian party system, the new regime of Abdelaziz Bouteflika, heavily
but not overwhelmingly influenced by the military (Mortimer, 2006), relied
both on the ruling party of the previous regime and on a new formation,
the National Democratic Rally, created by the military in the midst of the
civil war. These parties jointly controlled the legislature, and both played
a role in the executive. A plethora of other parties, some of them overtly
pro-regime and others representing moderate opposition, were also visible
in the electoral arena. This rather fractious party system has demonstrated
its long-term sustainability by surviving the demise of Bouteflika’s personal
rule in 2019 (Volpi, 2020). Three authoritarian party systems masterminded
by military regimes — those of Gambia, Mauritania, and Sudan — were
all characterized by outright domination of the ruling parties, with the
permitted opposition playing largely symbolic roles. All these party sys-
tems collapsed after military coups and/or revolutions that terminated the
respective political regimes.
The authoritarian party systems centered around political-military
groups that came to power as a result of violent civil conflicts, while
structurally similar to the party systems in the previous category, have
displayed much higher rates of long-term survival. The Ethiopian People’s
Revolutionary Democratic Front came to power as a coalition of ethnic
political-military groups that joined efforts to overthrow the previous dicta-
torial regime. From 1995 through 2015, the ruling party of Ethiopia contin-
ued to use ethnic political mobilization to win vast majorities of seats in the
b4407_Ch03.indd 46 1/6/2022 9:15:03 AM

country’s legislature (Abbink, 2011). The remaining seats were distributed

mostly among parties that were also ethnic-based and normally supported
the policies of the regime but, for a variety of tactical reasons, remained
unabsorbed by the ruling party. The Patriotic Salvation Movement came to
power in Chad when the forces of its leader, Idriss Déby, took over the capital
of the country after winning a civil war in 1990. Starting in 1996, multiparty
elections were not only allowed but also contested by a number of groups
representing those societal and/or territorial clusters of the country that did
not fall under the direct military control of Déby (van Dijk, 2005). Thus,
Chadian elections, albeit marred by electoral fraud and invariably resulting
in victories of Déby and his party, did endow the anti-regime parties with
some more than symbolic legislative representation.
This is not the case with the party systems that emerged after violent civil
conflicts in three adjacent countries, Burundi, Rwanda, and Uganda. In the
first two countries, the absolute domination of the ruling parties formed on
the basis of winning military-political organizations has remained unchal-
lenged by weak and toothless opposition parties permitted by the authorities
(Vandeginste, 2014). In Uganda, after a long period of experimentation with
“non-party democracy” that was effectively equivalent to the single-party
rule of the National Resistance Movement (Sejjaaka, 2004), multiparty
elections were restored in 2006 and contested by a number of parties with
relatively convincing opposition credentials. The most successful among
these parties, the Forum for Democratic Change, was created mostly by the
former members of the ruling party. In this relatively competitive context,
the political domination of the ruling party remained sustainable as a result
of widespread election violence and fraud (Helle and Rakner, 2014).

Post-Communist Countries
Most of the East European countries successfully democratized after the
fall of the communist regimes. The only exception outside of the former
b4407_Ch03.indd 47 1/6/2022 9:15:03 AM

6"×9"
Soviet Union is Serbia where, for nearly a decade, President Slobodan

Milosevic presided over a political regime in which legislative support to
his nationalist policies was provided a communist successor party, renamed
as the Socialist Party of Serbia, while the nationalist Serbian Radical Party
performed as the second-largest party. Pro-democracy opposition parties
also ran in elections, not entirely unsuccessfully, but they were unable to
unseat the regime that combined its reliance on the nationalist sentiment
in the electorate with outright election fraud (Bieber, 2003). After Serbia’s
poor performance in the 1997–1999 war over Kosovo, the nationalist pro-
Milosevic sentiment largely vanished, while his attempt to stay in power by
fixing elections caused massive unrest that resulted in the 2000 revolution
and subsequent democratization (Vladisavljević, 2016).
All other post-communist authoritarian party systems emerged in the
countries of the former Soviet Union. The earliest among them came into
existence in Azerbaijan in 1995, soon after the former communist leader
of Azerbaijan, Heydar Aliyev, returned to power. Naturally enough, the
apparatus of the Communist party was instrumental in building the New
Azerbaijan Party created by Heydar Aliyev and later inherited, together with
the presidency, by his son Ilham Aliyev. While the New Azerbaijan Party has
continuously relied on the support of many pro-regime independents on the
legislative floor, thus never formally winning a legislative majority, the repre-
sentation of the fractious and not very credible opposition was all but symbolic
(LaPorte, 2015). This was not the case in neighboring Armenia where the
Republican Party, rooted in the 1990–1991 struggle for independence, was
able to win only thin legislative majorities after coming to power in 1999. The
political legacies of the brief democratization experienced by Armenia in the
1990s, in the form of a viable opposition, remained in place in the course of
the authoritarian interlude (Iskandaryan, 2018). In a turn of events similar
to what happened in Serbia in 2000, the regime of the Republican Party fell
in 2018 amidst massive protests caused by allegations of election fraud.
The case of Russia is important not only for itself, but also because
of the profound influence exerted by Russia on several other post-Soviet
b4407_Ch03.indd 48 1/6/2022 9:15:03 AM

countries (Roberts, 2015). After a period of experimentation with democ-

racy that yielded a fractious and unstable party system (Golosov, 2003b), the
consolidation of power in the hands of president Vladimir Putin brought
his party, United Russia, to the core of the emerging authoritarian party
system. The two main parties inherited from the 1990s, as well as several
new formations, found their place on the margins of this system in the roles
of loyal opposition or vote splitters (Golosov, 2006b). Their representation,
albeit never negligible, was small enough to allow for United Russia’s full
control over the country’s legislature. To a large and increasing extent, the
electoral dominance of United Russia stems from its privileged access to the
media, political restrictions, and outright election fraud (Golosov, 2014d).
The dominant party of Kazakhstan, known for the most part of its
history as Nur Otan (Radiant Fatherland), was formed in 1999 as the main
political vehicle of president Nursultan Nazarbayev, originally by emulating
one of Russia’s leading parties of the time, Fatherland. Thus, the similarities
between the two authoritarian party systems noticed in the literature (Isaacs
and Whitmore, 2014) are not entirely typological. Unlike Russia, however,
Kazakhstan did not pass through a democratic phase of post-communist
transition. Throughout the 1990s, Nazarbayev’s control over the national
legislature was provided by pro-regime independents (Satpaev, 2007). In
the legislatures elected in 1999 through 2004, Nur Otan was accompanied
by a relatively large faction of the loyal opposition, mostly in the form of
the Civic Party of Kazakhstan (Del Sordi, 2016). Later, however, it merged
into the ruling party. Starting with 2007, the shares of seats jointly held by
opposition parties varied from zero to 15 percent at best.
In Tajikistan, the 1992–1997 civil war, in the process of which president
Emomali Rahmon consolidated his power and established a dictatorial
regime (Lynch, 2001), produced a number of military-political groups
representing different regions of the conflict-ridden country (Dudoignon,
1997), but most of them did not survive into the post-war phase. Starting
from 2000, Rahmon has derived legislative support mostly from the People’s
Democratic Party that absorbed the organizational residua of the Communist
b4407_Ch03.indd 49 1/6/2022 9:15:03 AM

6"×9"
party apparatus. Several other parties enjoy rather cosy relations with the
authorities, thus representing only token opposition to the Rahmon regime.
A similar process of domestication, greatly facilitated by systematic harass-
ment, had been experienced by the once-powerful at the regional level
Islamic Renaissance Party (Epkenhans, 2018), but eventually it was banned
by the authorities.
The rule of Islam Karimov in Uzbekistan (1991–2016) in certain ways —
particularly, in developing a state ideology, non-compliance with which led
to political exclusion (March, 2003) — bore striking similarity to its Soviet
predecessor. The earliest of the pro-Karimov political parties, the People’s
Democratic Party of Uzbekistan, emerged as a successor to the local branch
of the past ruling party, but it was never elevated to the dominant position
enjoyed by the pro-regime parties of other post-Soviet states. Instead, signif-
icant shares of legislative seats were continuously gained by several parties
masterminded by the authorities and claiming to represent different societal
constituencies (Sirojjon, 2015), such as the “pro-business” Uzbekistan Liberal
Democratic Party. All official parties of Uzbekistan pledged their loyalty to
Karimov and his official ideology. The system survived after the death of
Karimov in 2016.
b4407_Ch03.indd 50 1/6/2022 9:15:03 AM

Chapter 4
The Properties of Authoritarian

Party Systems
T
his study draws evidence from the whole universe of party systems
that are empirically observable in the conditions of authoritarian
regimes. As shown in Chapter 3, the phenomena under investi-
gation are very diverse, which naturally increases the dangers of “concept
misformation,” as it was labeled in an influential article of Sartori (1970)
and noticed by many authors of early theoretical diatribes against the
comparative/quantitative direction in political science (MacIntyre, 1971).
In his seminal treatment of the comparative method, Lijphart (1971)
identifies several ways in which political science can overcome these
dangers without abandoning the very strategy of wide cross-national
comparisons. One of these ways is to identify a limited number of com-
parable parameters. Such parameters are related to the essential properties
of these phenomena, i.e., those properties that can be attributed to all of
them across the whole spectrum of available observations. For the pur-
poses of quantitative research, it is highly desirable that these properties
are measurable.
This chapter is focused on four such properties of party systems:
fragmentation, format, volatility, and nationalization. Many more properties
have been identified in the literature, especially in in-depth country case
studies (Niedermayer, 1995), which is only natural because party systems
are complex, multidimensional phenomena (Gross and Sigelman, 1984,
p. 463). Ascribing some of these properties to authoritarian party systems
does indeed take a good deal of concept misformation, which is the case
51
b4407_Ch04.indd 51 1/6/2022 9:15:14 AM

6"×9"
with such a fundamental and measurable property as ideological polariza-

tion (Lachat, 2008). Of course, authoritarian party systems may include
parties with different ideological positions, and the resulting distances
among these parties are measurable by using some of the methodological
tools that are common in cross-national research on democracies (Basedau
and Stroh, 2011). But the controlled nature of political oppositions, when
seemingly radical and even extreme parties, such as the communists and
nationalists, can be supportive of the government of a very different ideo-
logical persuasion and peacefully reside in the electoral niches prescribed by
the authorities, makes cross-national variation on this parameter less than
relevant. Yet another important property of party systems, institutionaliza-
tion (Mainwaring and Scully, 1995), is not entirely relevant for this study
because all of the included party systems satisfy the minimum criterion of
sustainability in the course of three consequent elections.
In contrast, cross-national variations on the four parameters listed above
are not only quantitatively significant, as it will be shown in the following,
but also substantively important for our understanding of different varieties
of authoritarian party systems. Moreover, they are instrumental in clearing
the empirical distinction between democratic and authoritarian party sys-
tems. Of course, no claim can be made that these two kinds of party systems
contain functionally equivalent elements, which would constitute a typical
functionalist fallacy long rejected in political science (Callinicos, 1987).
What can be empirically registered between them is structural isomorphism,
commonly understood as “a constraining process that forces one unit in a
population to resemble other units that face the same set of environmental
conditions” (DiMaggio and Powell, 1983, p. 149). In this particular case,
isomorphism is generated by a set of constraints that are inherent in the
conditions of electoral competition, be it free and fair, as in democracies,
or limited and unfair, as in authoritarian regimes. This is why single-party
systems, with their lack of party-structured elections, are not included in
this analysis. In such systems, the parameters of interest are constant. But
electoral authoritarian regimes, exactly because they allow for inter-party
b4407_Ch04.indd 52 1/6/2022 9:15:14 AM

The Properties of Authoritarian Party Systems 53
competition, are eligible for comparisons with other party systems, both
authoritarian and democratic.
One of the problems that have to be solved in this chapter is the prob-
lem of measurement. Even if the set of essential party system properties can
be identified with a large degree of certainty, the tools of measuring these
properties need to be sharpened, not only for their general improvement
but also, particularly, for a more productive use in research on authoritarian
party systems. This applies to the measurement of all the four party system
properties that are central in this study, but some of the conventional tools,
such as the effective number of parties as a primary instrument for measuring
party system fragmentation, deserve extensive treatment. This explains why
a large portion of this chapter deals with methodological issues. My goal is
to show empirically how authoritarian party systems vary among themselves
and differ from democratic party systems on the parameters of substantive
importance, so that it will be possible to build explanatory models and to
test them empirically in the subsequent chapters. Therefore, the choice of
the tools of measurement, as long as this choice is problematic, has to be
explained in detail.
Party System Fragmentation

Starting with the path-breaking work of Duverger (1954), fragmentation,
also referred to as fractionalization, has been always viewed as the most
important party system property and, in this capacity, as a major basis for the
classification of party systems (Smith, 1989). This approach draws a major
distinction between two-party and multiparty systems, which has been for
a long time recognized as a theoretically fundamental divide (Daalder, 2002,
pp. 43–51). The conceptual loading of the notion of fragmentation and its
opposite, concentration, is very simple. Highly fragmented party systems
include many “viable,” or “important,” or (to put it most radically) “real”
political parties in a party system that comprises parties of unequal sizes,
while low fragmentation is associated with the presence of few such parties.
b4407_Ch04.indd 53 1/6/2022 9:15:14 AM

6"×9"
However, this simple concept entails a complex methodological problem.

Which parties should be counted as “important?”
In comparative political party research, we need to distinguish between
a four-party constellation with party vote shares (0.52, 0.45, 0.02, 0.01),
which is apparently a two-party system, and a four-party constellation with
party vote shares (0.40, 0.25, 0.20, 0.15), which is a case of multipartism.
There should be a way to discount very small parties. The methodological
problem is that no such way is readily available. Of course, it is possible to
set an arbitrary threshold of exclusion, discounting all parties that fail to
reach one, or three, or five percent of the vote (Ware, 1995). Yet, it is clear
that a five percent party might be unimportant if other parties enjoy massive
support, and it might be quite important if other parties are comparably
weak — for instance, if the largest party’s vote share is 15 percent. Then we
need to quantify the idea of the number of parties in a systematic way that
allows for taking into account the relative sizes of parties, which is impossible
without using a measure expressed as a mathematical formula.
The first attempt at measuring party system fragmentation was related
to a much earlier experience in a different disciplinary field, economics. The
Herfindahl–Hirschman index of industrial concentration, developed in the
1940s, is widely used in applied economics, and it is currently employed as
a standard measure of industrial/market concentration in the US economic
statistics (Warren-Boulton, 1990; see also Taagepera and Grofman, 1981).
The formula for the Herfindahl–Hirschman index is thus:
x
HH = ∑ si , (4.1)

where si is the size of the ith component expressed as its decimal (absolute)
share of the total, and sigma stands for summation.
A very useful property of the Herfindahl–Hirschman index, as well
as of all indices based on it, is exactly that it allows for excluding from our
calculus the actual number of components. In cross-national research using
aggregate election results from the past — when minor parties were often
b4407_Ch04.indd 54 1/6/2022 9:15:14 AM

lumped together under the heading “others” — this widens the scope of
empirical inquiry. The index is also a probability measure. This was noticed
by Rae (1967, pp. 55–56). When introducing his index of party system
fractionalization:
F = 1 − HH, (4.2)
he emphasized that it represented “the frequency with which pairs of voters

would disagree [in their choice of parties] if an entire electorate interacted
randomly.” An advantage shared by the two indices is that the values of both
indices run from 0 to 1, which is often considered as a desirable property
for indices used in statistical analysis.
The index of Rae was the first but not the last attempt to utilize the
Herfindahl–Hirschman index in comparative research on party systems. In
fact, both indices were decisively squeezed out of circulation after Laakso and
Taagepera (1979) proposed their effective number of parties, defined as follows:
1
N LT = x
. (4.3)
∑s 2
i
1
A useful property of the Laakso–Taagepera index is that it can be cal-
culated not only on the basis of the decimal shares of components, as in all
formulas above, but also on the basis of the raw numbers of votes or seats
received by each of the parties. In order to achieve that, we have to use a
different formula:
2
x 
∑ si 
= x  ,
 1
N LT (4.4)
∑ si2
1
where si is the number of votes or the number of seats received by the ith party.
This formula yields values equivalent to those yielded by the previous one,
yet it avoids minor deviations stemming from rounding.
b4407_Ch04.indd 55 1/6/2022 9:15:14 AM

6"×9"
The crucial parameter on which the Laakso–Taagepera index outper-

forms its predecessors is that it has more intuitive content. Unlike earlier
indices, it can be interpreted as the number of equal-size parties to which the
given constellation of unequal-size parties is equivalent. As put by Taagepera
(2002, p. 5), “N[LT] = 3.2 evokes the image of approximately three parties,
while the corresponding HH = 0.31 or F = 0.69 evoke no such images… The
latter are abstract indices, while N[LT] can be visualized.” Indeed, in most cases
the visualizations provided by the index do not fail to meet our intuitive
expectations. The major properties of the index that make it instrumental
as a measure of the effective number of parties are thus: First, and most
important, if all components in the given party constellation are equal, the
value of the index equals the actual number of components. Second, if the
components are not equal, the values of the index are smaller than the actual
number of components. Third, if any of the components in the given party
constellation becomes smaller, so that the actual number of components
goes up, the value of the Laakso–Taagepera index also goes up. For instance,
the value of the index for a four-party constellation (0.52, 0.45, 0.02, 0.01),
2.11, is greater than for a three-party constellation (0.52, 0.46, 0.02), 2.08.
These three properties are necessary for any index that is intended to
measure the effective number of parties. Therefore, they can be viewed as
basic requirements that have to be fulfilled by such indices. In what follows,
for the sake of brevity, I will occasionally refer to them as the first, second,
and third basic requirements.
Measures of the effective number of parties are single-number indices,
and it is natural for them to yield equal values for different party constel-
lations. This problem, if it is a problem at all, has no solution. What is
important for the measures of the effective number of parties is to produce
values that are close to our intuitive expectations, which can and should be
provided by index builders. From this perspective, the major discrepancy
between the values of the Laakso–Taagepera index and our intuitive expec-
tations regarding the effective number of parties was noticed already by
the first scholar who proposed an alternative to the Laakso–Taagepera
b4407_Ch04.indd 56 1/6/2022 9:15:14 AM

index: Molinar (1991). He was particularly concerned with party con-

stellations in which the share of the largest component, s1, is greater than
50 percent. Already at s1 = 0.8, the lower limit value of the Laakso–Taagepera
index is 1.47, and at s1 = 0.67, it is 1.79. Most people would not agree to
consider a situation in which the largest party takes two-thirds of the vote
as a nearly two-party constellation, but this is what the value of the Laakso–
Taagepera index suggests. Moreover, its upper value at s1 = 0.67 is 2.25. Here,
a two-party system is registered for a constellation that includes one party
that receives two-thirds of the vote, and an infinite number of extremely
small parties. In such situations, the Laakso–Taagepera index obviously
fails on intuitive content.
Unfortunately, the alternative formula suggested by Molinar (1991,
pp. 1384–1385) did not solve the problem. A very detailed and well-substan-
tiated critique of the Molinar index was thoroughly presented by Dunleavy
and Boucek (2003, pp. 308–313). To put it in brief, the index violates the
third basic requirement at approximately s1 > 0.35. It is not surprising, then,
that it registers very low levels of party system fragmentation at s1 > 0.5, thus
apparently solving the major problem of the Laakso–Taagepera index. Yet,
the price attached to this solution is too high: most of us will not recognize
the party constellation (0.6, 0.4) as more fragmented than (0.6, 0.2, 0.2).
In practice, the Molinar index was very rarely used in political science
research. But the effort of Molinar probably contributed to a wider recog-
nition of the major problem of the Laakso–Taagepera index. Taagepera
(1999) recommended to use 1/s1 as a supplement to the effective number
of parties, especially in those cases when it conceals the dominance of the
largest party (the example of India in 1952–1984 is cited). There were also
several attempts to develop indices that meet specialized research needs,
such as an index that replaces proportions of seats in the calculus for the
Laakso–Taagepera index with the Banzhaf voting power measures (Dumont
and Caulier, 2003), which is useful for the study of coalition politics, and an
index that counts as “viable” those candidates who cross a specified quota
(Niemi and Hsieh, 2002), which is useful for the study of certain aspects
b4407_Ch04.indd 57 1/6/2022 9:15:14 AM

6"×9"
of district elections. It is clear that such indices cannot be used for general
purposes because they do not satisfy the basic requirements to the general
measures of the effective number of parties. Yet noticeably, all the proposed
supplements and special-need alternatives to the Laakso–Taagepera index
deal primarily with the problem identified above as its major deficiency, the
inability to adequately represent cases of one-party dominance.
In order to remedy this shortcoming, Taagepera (1999) proposed to use
the inverse of the largest component, 1/s1, as a supplementary measure. He
was careful enough to dismiss the idea that 1/s1 could be used as a component
for creating a new composite index, rather than as a mere supplement to it:
“This is about as wishful as hoping to combine the mean and the standard
deviation of a distribution into a single measure” (Taagepera, 1999, p. 503).
The index proposed by Dunleavy and Boucek (2003), who rejoined that “two
inherently limited numbers are not much use to anyone, and if they can be
combined more productively then they should be” (p. 302), is simply the
average of the Laakso–Taagepera index and 1/s1:
 1 1 1
NB = + × . (4.5)
 x 2 s 2
 ∑ si i

1
The Dunleavy–Boucek index fully satisfies all three basic requirements
to the measures of the effective number of parties. At the low levels of frag-
mentation, the values of the index are intuitively plausible. But the Laakso–
Taagepera index fares better at expressing high levels of fragmentation.
An alternative index, primarily with the aim of achieving greater cor-
respondence between the values of the effective number of parties and our
intuitive expectations, has been proposed by Golosov (2010). Consider
constellation (0.8, 0.2). How many “effective” parties are there? Should we
register 1.47 parties (Laakso–Taagepera) or 1.36 parties (Dunleavy–Boucek)?
Both numbers are abstract. In fact, our intuition tells a different story: in
this constellation, we have a large party, and there are all reasons to count
it as one, as argued already by Molinar (1991, p. 1385), and a party that is
b4407_Ch04.indd 58 1/6/2022 9:15:14 AM

one-fourth the size of the large party. Then the effective number of parties
should be 1.25. This solution is not only intuitively appealing but also logi-
cally unavoidable, especially if we think of the effective number of parties in
terms of the equivalent number of equal-size components. The formula that
yields this value, 1/s1, is already at our disposal. The task is to build an alge-
braic expression reducible to this formula when applied to equal-component
constellations, for which it will be equivalent to the actual number, and also
when applied to two-component constellations with s1 > 0.5, for which it will
be equivalent to 1 + (s2/s1). For example, Barnea and Rahat (2007, p. 384)
introduce the expression s2/s1 as an ad hoc competitiveness index.
This goal cannot be achieved if we continue to use the Herfindahl–
Hirschman index as the computational core of the effective number of
parties, which is the case with the Laakso–Taagepera, Dunleavy–Boucek,
and Molinar indices. The Golosov index is algebraically expressed as follows:
x si
Np = ∑ , (4.6)
1 si + s12 − si2

or, when dealing with constellations without zero-size components, a trans-
formed expression can be used:
x
1
Np = ∑ . (4.7)
1 1 + (s12 / si ) − si

Like the Laakso–Taagepera index, the Golosov index can be calculated
not only on the basis of the decimal shares of components, as in the two
formulas above, but also on the basis of the raw numbers of votes or seats
received by each of the parties. For this, we have to use the formula that is
derivative from the previous one:
x
x ∑ si
Np = ∑ x
1
. (4.8)
1
∑ si + (s12 / si ) − si
1
b4407_Ch04.indd 59 1/6/2022 9:15:15 AM

6"×9"
x
This may seem complex but note that ∑1 s i is simply the overall num-
ber of votes cast or seats distributed in the given election, a statistic that is
normally available without additional calculations.
The Golosov index satisfies all three basic requirements to the measures
of fragmentation. When applied to two-component constellations with s1 >
0.5, it does yield values equivalent to 1 + (s2/s1). As a result, most importantly
for this study, its values for constellations with low levels of fragmentation
are intuitively plausible, but this does not entail counter-intuitively low val-
ues for those constellations when fragmentation is high. For instance, the
values of Laakso–Taagepera, Dunleavy–Boucek, and Golosov indices for
the extremely fragmented constellation with s1 = 0.2 and 80 components at
0.01 each are 20.83, 12.91, and 17.03, respectively. For illustrative purposes,
Table 4.1 reports the values of three indices for eight hypothetical vote or
seat constellations. Four of these constellations represent minimum levels
of fragmentation at any given s1 (A, C, E, and G). Four other constellations,
while not reaching the maximum levels of fragmentation at any given s1,
are nevertheless very fragmented, because each of them sets s2 at 0.1 and
includes 15 components at 0.01 (B, D, F, and H).
Table 4.1. The values of different measures of fragmentation for eight

hypothetical vote or seat constellations.
Largest
component,
fractional Other components, Laakso– Dunleavy–
Constellation share fractional shares Taagepera Boucek Golosov
A 0.75 0.25 1.60 1.47 1.33
B 0.75 0.1, 15 at 0.01 1.74 1.54 1.42
C 0.55 0.45 1.98 1.90 1.82
D 0.55 3 at 0.1, 15 at 0.01 2.99 2.41 2.24
E 0.35 0.35, 0.3 2.99 2.92 2.90
F 0.35 5 at 0.1, 15 at 0.01 5.75 4.30 4.49
G 0.15 5 at 0.15, 0.1 6.90 6.78 6.89
H 0.15 7 at 0.1, 15 at 0.01 10.64 8.65 11.85
b4407_Ch04.indd 60 1/6/2022 9:15:15 AM

As is evident from the table, the values of the Laakso–Taagepera and

Golosov indices for constellations with moderate levels of fragmentation,
which is typically the case in democracies, are very similar, so that the two
indices can be used interchangeably. Thus, the long pedigree of the Laakso–
Taagepera index in quantitative political science explains its continued
prevalence in the studies of democracies. The party systems of authoritarian
regimes are often highly concentrated, as a result of which the Golosov index
can be viewed as the instrument of first preference. This is the approach taken
in this book. In the further analysis, the term “effective number of parties”
is used exclusively to refer to the values yielded by the Golosov index.
The effective number of parties can be established in two ways, for vote
distributions and seat distributions, which is reflected, respectively, in the
measures of electoral fragmentation and legislative fragmentation. I focus
on the latter. One of the reasons for this choice is merely technical. In the
1940s–1980s, the state of election statistics reporting in many countries,
but especially in some of the authoritarian regimes, was very poor. Vote
distributions for many elections are simply not available. For a long time,
it was conventional to use seat distributions as an available alternative even
in research on democracies. When taking this approach, I am completely
aware that the effective numbers of electoral and legislative parties are never
the same, with the difference stemming mainly from the effects of elec-
toral systems and, to a much lesser extent, from a number of other factors
(Golosov, 2017). In some research contexts, these differences should not be
discounted. But the context of this research is different.
Authoritarian regimes use their party systems primarily for the purpose
of exerting control over legislatures. From this perspective, it is legislative
fragmentation that defines the modus operandi of an authoritarian party
system by establishing a specific relationship between the dictatorial exec-
utive, the pro-executive party or parties, and the permitted opposition. At
the same time, electoral fragmentation, while directly relevant to voting
behavior, is only indirectly connected with the functioning of authoritar-
ian regimes because its political consequences can be largely neutralized
b4407_Ch04.indd 61 1/6/2022 9:15:15 AM

6"×9"
by institutional means, such as the choice of a restrictive electoral system,

or by political means. If, however, an authoritarian regime chooses not
to neutralize these consequences, then high fragmentation is compatible
with the properties of the authoritarian party system as generated by the
given political or institutional variety of dictatorial rule. Since the political
regime and institutional determinants of authoritarian party systems are of
primary interest for this study, its substantive focus justifies my reliance on
the measures of legislative fragmentation.
One of the standard problems in the calculation of the effective number
of parties is related to the non-party (independent) candidates or elected
representatives. Possible solutions to this problem include counting each
of them as an individual party, which may lead to vastly exaggerated esti-
mates of fragmentation, or, alternatively, counting them all as a single party
(Herron and Nishikawa, 2001). When dealing with authoritarian party
systems, the problem is further complicated by the fact that in some cases,
non-party politicians tend to closely cooperate with the pro-regime parties
(Lust-Okar, 2006), but in other cases, such politicians run as independents
simply because their parties are not legally eligible for electoral participation
even if effectively they do provide inter-candidate coordination (Ghanem
and Mustafa, 2011). In order to solve this problem, I employed a three-step
procedure. First, I calculated the effective number of parties by counting all
independents as a single party. Second, I performed calculations by summing
up the shares of seats gained by the largest party and all independents. Third,
I averaged the values yielded by the previous calculations. These averages
were used for all quantitative analyses reported in this book.
In the analysis below, the empirical basis for comparisons
among authoritarian party systems is provided by the dataset described in
Chapter 2 of this book. For drawing comparisons between authoritarian and
democratic party systems, it is necessary to create a separate dataset covering
democracies. This has been achieved by creating a selection of cases on a
country/year basis. The criteria for case selection are thus: (1) maximizing
the diversity of the included cases in terms of cultural backgrounds, levels
b4407_Ch04.indd 62 1/6/2022 9:15:15 AM

of socio-economic development, and longevity of democracy, so that the

diversity of authoritarian party systems could be matched; (2) including only
undeniably democratic countries that scored 1 or 2 on the political rights
scale of the Freedom House at the time of holding the selected elections;
(3) including only countries with party-structured elections; (4) including
mostly those elections that can be viewed as manifestations of relatively
stable, institutionalized party systems; and (5) satisfying the particularly
high data demands of the measurement of party system nationalization.
Only elections to the single or lower chambers of national legislatures are
included. Provided that in this book the study of democratic party systems
is undertaken mostly for illustrative purposes, this procedure seems to be
sufficiently rigorous.
Each of the selected 74 country cases enters the dataset as a sequence
of two elections held from 1987 to 2014. For each of the cases, the mea-
surement of fragmentation and format was performed on the basis of both
observations, which results in 148 units overall. The measurement of vola-
tility and nationalization was performed only for the most recent elections
in each of the sequences. The country-year composition of the dataset
is thus: Argentina 1997–1999; Australia 2004–2007; Austria 2002–2006;
Bahamas 2007–2012; Barbados 2008–2013; Belgium 2007–2010; Belize
2008–2012; Bolivia 1989–1993; Brazil 2002–2006; Bulgaria 1994–1997;
Cabo Verde 1995–2001; Canada 2006–2008; Chile 2005–2009; Costa Rica
2002–2006; Croatia 2003–2007; Cyprus 2006–2011; Czech Republic 2002–
2006; Denmark 2007–2011; Dominica 2005–2009; Dominican Republic
2006–2010; El Salvador 2009–2012; Estonia 1999–2003; Finland 2007–2011;
France 2007–2012; Ghana 2004–2008; Greece 2000–2004; Grenada 2003–
2008; Guyana 2001–2006; Hungary 1990–1994; Iceland 2007–2009; India
1999–2004; Indonesia 2009–2014; Ireland 2002–2007; Israel 1999–2003; Italy
1987–1992; Jamaica 2007–2011; Japan 2005–2009; Liechtenstein 2001–2005;
Luxembourg 2004–2009; Malta 2003–2008; Mauritius 2005–2010; Mexico
2003–2006; Mongolia 2000–2004; Namibia 1999–2004; Netherlands 2006–
2010; New Zealand 2005–2008; Norway 2005–2009; Panama 2004–2009;
b4407_Ch04.indd 63 1/6/2022 9:15:15 AM

6"×9"
Peru 2006–2011; Philippines 1995–1998; Poland 2007–2011; Portugal

2005–2009; Romania 1996–2000; St. Kitts and Nevis 2004–2010; St. Lucia
2006–2011; St. Vincent and Grenadines 2005–2010; Samoa 1996–2001; San
Marino 2006–2008; Sao Tome and Principe 1991–1994; Senegal 2007–2012;
Serbia 2008–2012; Sierra Leone 2007–2012; Slovenia 2004–2008; South
Africa 2004–2009; South Korea 1996–2000; Spain 2000–2004; Suriname
2005–2010; Sweden 2006–2010; Switzerland 2007–2011; Taiwan 2001–2004;
Trinidad and Tobago 2007–2010; United Kingdom 2005–2010; United States
2004–2006; Uruguay 2004–2009.
A large part of the data used for the calculation of the indices presented
in this chapter has been derived from a series of fundamental publications of
Dieter Nohlen and his colleagues (Nohlen et al., 1999; Nohlen et al., 2001a,
2001b; Nohlen, 2005a, 2005b; Nohlen and Stöver, 2010). Information about
the results of recent elections that are not covered in these publications has
been mostly derived from the PARLINE database of the Inter-Parliamentary
Union (http://data.ipu.org and http://archive.ipu.org/parline-e/parline
search.asp, both accessed at different times in 2009–2021). In certain cases,
I consulted national election statistics sources.
Table 4.2 shows the aggregate values of the effective number of parties
for several categories of party systems. Tables 4.3 and 4.4 show the results
of the related statistical analysis. In order to provide for a deeper under-
standing of the interconnection between the level of democracy and party
Table 4.2. Fragmentation of democratic and authoritarian party systems

(effective number of parties).
Standard
Mean Median Minimum Maximum deviation N
Democratic party systems 1 2.96 2.46 1.17 8.19 1.50 96
Authoritarian party systems 1.77 1.39 1.00 11.80 1.32 294
by election
by system
b4407_Ch04.indd 64 1/6/2022 9:15:15 AM

Table 4.3. Summary of ANOVA comparison of party systems by

fragmentation.
Sum of Degrees of
squares freedom Mean square F Significance
Between groups 56.67 2 28.34 12.63 0.00
Within groups 453.10 202 2.24 — —
Total 509.77 204 — — —

authoritarian party systems by fragmentation.
Mean Standard
difference error Significance
Democratic party Democratic party systems 2 0.02 0.26 1.00
systems 1 Authoritarian party systems 1.18 0.25 0.00
Democratic party Democratic party systems 1 −0.02 0.26 1.00
systems 2 Authoritarian party systems 1.16 0.29 0.00
Authoritarian Democratic party systems 1 −1.18 0.25 0.00
party systems Democratic party systems 2 −1.16 0.29 0.00
system fragmentation, I divide the group of democracies into two subgroups,

those with the Freedom House political rights scores of 1 and 2 at the time
of holding elections. They are referred to as democratic party systems 1 and
democratic party systems 2, respectively. The fragmentation of authoritarian
party systems is calculated with the use of two different algorithms. The
line of the table labeled “authoritarian party systems by election” reports
aggregate statistics on the basis of individual elections. For the line labeled
“authoritarian party systems by system,” I first calculated averages within
individual party system categories, and then aggregated them. Calculations
for Tables 4.3 and 4.4 have been performed on the basis of system-level
indicators for authoritarian party systems.
As follows from the table, authoritarian party systems are indeed deci-
sively less fragmented than those of democracies. This differentiates author-
itarian party systems not only from the party systems of well-established
b4407_Ch04.indd 65 1/6/2022 9:15:15 AM

6"×9"
democracies that can be found primarily among the countries with the
highest political rights score, but also from the more uncertain polities that
prevail in the category of democratic party systems 2. At the same time, it
is noticeable that extremely high levels of fragmentation are not alien to
authoritarian party systems, as evident from the fact that the greatest value
in the table is registered for the results of one authoritarian election. The
determinants of the observed variation will be discussed at length in the
subsequent chapters of this book.
Party System Format

The concept of party format is related to the concept of fragmentation to an
extent that they are sometimes, even if implicitly, viewed as identical. But
they are not. In this study, I use the term “party system format” to refer to
the number of parties in its relationship to the possible patterns of coalition
politics. Consider two-party systems, one with the seat distribution of (0.48,
0.46, 0.06), and the other with the seat distribution of (0.51, 0.43, 0.06). The
effective numbers of parties in these systems are nearly identical if measured
with the Laakso–Taagepera index (2.24 and 2.23, respectively). The Golosov
index, by attributing them the respective values of 2.17 and 2.04, does regis-
ter some difference, but it is insignificant. Meanwhile, the political distance
between the two systems is colossal. The first of them makes it impossible
for the leading party to rule alone. It is necessary to build a coalition, which
is obviously not the case with the second system. This difference cannot
be registered by any general purpose measure of fragmentation, however
mathematically constructed, which made some prominent scholars such as
Sartori (1986) quite skeptical regarding the quantitative measures of party
system fragmentation.
I do not share such skepticism. The effective number of parties makes
it possible to distinguish between a system with three important parties
and a system with a dozen of them, which certainly makes a difference.
But the problem is real and for a long time — in fact, well in advance of the
b4407_Ch04.indd 66 1/6/2022 9:15:15 AM

earliest attempts of party system measurement — it has been addressed in

political science by building ordinal classifications of party systems based
on the numbers of parties but putting emphasis on the patterns of coali-
tion politics. Of course, the distinction between two-party systems and
multiparty systems is fundamental. It is central for several salient streams
in political science research, including the interplay between electoral sys-
tems and party systems (Riker, 1982), government formation and duration
(Grofman, 1989), the basic properties of different institutional designs
(Shugart and Carey, 1992), and government performance (Chhibber and
Nooruddin, 2004).
Duverger (1954) made a decisive step toward conceptualizing the
profound difference between two-party and multiparty systems that had
been noticed by political scientists and practitioners as long ago as in the
19th century (Riker, 1982). The second important step was made by Blondel
(1968). His contribution was not only to define the formats identified by
Duverger in operational terms, but also to introduce a number of additional
categories. The basis for Blondel’s classification is the share of the vote jointly
received by two largest parties in the legislative election. Two-party systems
are those in which it exceeds 90 percent. For those systems where it makes
75 percent of the vote or more, Blondel introduces a new type of two-and-
a-half party systems. All other systems are genuinely multiparty, but they
are divided into two types, those with and without a dominant party. These
types are distinguished by the share of the vote solely obtained by the largest
party, the threshold being set at 40 percent.
Rokkan (1970) produced a very similar list of categories but introduced
two major methodological improvements: first, Rokkan shifts the basis for
classification from vote shares to seat shares, and second, his approach is
centered not so much on the absolute strengths of parties as on their relative
strengths. Both innovations were consistent with the substantive content of
the notion of party system format. It is also noticeable that Rokkan iden-
tifies two distinct forms of multipartism. In the Scandinavian “1 vs. 3–4”
system, the largest party either rules alone or alternates with a more or less
b4407_Ch04.indd 67 1/6/2022 9:15:15 AM

6"×9"
formalized coalition of minority parties. The even multiparty systems, “1 vs.

1 vs. 1 +2–3,” are the home for full-fledged, multilateral coalition politics.
Some of the ideas embedded in Rokkan’s approach received continu-
ation in the seminal work of Sartori (1976). From his perspective, parties
are relevant only if, by merit of their relative strengths on the legislative
floor, they can influence coalition formation. Such a potential can be either
positive, in the form of being acceptable as coalition partners (coalition
potential properly), or negative, in the form of being able to undermine
coalition formation (blackmail potential). The direction of the potential is
defined by the ideological distances among parties. This brings Sartori back
to the two-component classification of Duverger, with the two-party and
multiparty systems being the principal formats, yet the second category is
broken into two subdivisions by the criterion of ideological distance, which
yields distinct categories of “moderate pluralism” and “polarized pluralism.”
One of the innovations in Sartori’s work was to identify a new type, predom-
inant party systems, characterized by continuous parliamentary domination
of one political party. It has to be mentioned that Sartori provides no solid
conceptual justification for singling out this particular format, noting instead
that “the predominant-party system actually is a more-than-one party system
in which rotation does not occur in fact” (Sartori, 1976, p. 173). Apparently,
here the basis for his classification is purely empirical.
While Sartori’s classification is often regarded as the most influential
(Mair, 2002, pp. 91–92), it has to be recognized that for those scholars whose
research requires a more differentiated view of party systems, and who need
not focus so heavily on ideological distances, it has little to offer. Modified
versions of Blondel’s taxonomy remain widespread. In the relatively recent
literature, a serious attempt to build on Blondel’s work has been undertaken
by Siaroff (2019) who identifies as many as eight different formats by setting
a number of numerical criteria on the basis of the numbers of seats gained by
the first- and second-largest parties, on the one hand, and the ratios of these
numbers, on the other hand. A simple yet elegant approach to the empirical
identification of different party system formats has been developed in the
b4407_Ch04.indd 68 1/6/2022 9:15:15 AM

form of the so-called Nagayama diagram, often referred to as the Nagayama

triangle, which is a two-dimensional graphical representation of the space
of electoral competition. It was used mostly for analyzing competition
in single-member districts and proved to be quite efficient (Reed, 2001;
Likhtenchtein and Yargomskaya, 2005).
The properties of the diagram are systematically explained elsewhere
(Grofman et al., 2004). Here, it is sufficient to say that the x-axis is used to
show the absolute vote share of the winning candidate, and the y-axis, that of
the second runner-up. Because she never takes more votes than the winner,
the feasible set of data points lies within a triangle bounded by the x-axis
and segments of the lines y = x and y = 1 − x. By dividing the triangle into
several segments, different patterns of competition can be identified. The
Nagayama diagram was developed specifically for investigating the patterns
of competition in single-member districts, but it can be used, and in fact it has
been used, in more general party system research, provided that vote shares
are replaced with seat shares (Taagepera and Allik, 2006). Yet, as Grofman
et al. (2004) correctly suggest, for the graphical presentation of this kind
to be efficient, it has to allocate equal regions to all theoretically important
types. Indeed, it is imperative that the distribution of data points across the
available space be defined only by the own properties of the data, not by the
construction of the graph. While Grofman et al. (2004) demonstrate that this
desideratum can be satisfied when studying district contests, this is not the
case when the Nagayama graph is used in research on party systems. Besides,
some of the sectors of the Nagayama graph even theoretically cannot contain
any data points at all, which is also an undesirable property.
The method of identifying party system formats developed by Golosov
(2011a) builds mostly on the Nagayama diagram while also being informed
by the number-based ordinal classifications of party systems and by the
effective number of parties. In particular, the mathematical form of the
Golosov index is such that the largest component is always counted as one,
while all other components are weighted against the size of the largest com-
ponent and their own sizes. Then, taking the size of the largest component
b4407_Ch04.indd 69 1/6/2022 9:15:15 AM

6"×9"
as a constant, we can build a diagram with x-coordinates expressing the

relative size of the second component, and y-coordinates, the relative size
of the third component. In other words, the coordinates themselves can be
indices, single numbers derived from more than one quantity.
Such a solution is very close to one of the operational criteria developed
by Siaroff (2019), the seat ratio between the largest and the second-largest
party. Of course, to make the index bounded, the inverse of the ratio has
to be taken. Then the x-coordinates can be defined as s2/s1. But this allows
neither for separating constellations with a majority party from all others,
nor for differentiating between such obviously different-type combinations
as (0.6, 0.2, 0.2) and (0.3–7 parties at 0.1). The optimal representation can
be achieved by bringing in an additional element, the share of seats jointly
received by all parties from the fourth-largest to the smallest, sr. Thus, the
sizes of the second and the third components are weighted not only by the
size of the largest component but also by this additional element, one of
the bounds being provided by entering sr both into the numerator and the
denominator of the equations:
s2 + sr
x= , (4.9)
s1 + sr

s3 + sr
y= . (4.10)
s1 + sr

Figure 4.1 shows the resulting graphical representation dubbed the

relative-size triangle. Similar to the Nagayama diagram, the feasible set of
data points lies within a triangle, the bounds of which are the x-axis (AC),
the x = 1 line (BC), and the y = x line (AB). This determines the shape of
the triangle. Unlike the Nagayama diagram, the graph does not differentiate
among the degrees of multipartism. Constellations of more than two equal-
size components are all represented by the same data point, B. I find it theo-
retically justifiable. Of the qualitative party system classifications discussed in
the first section of this chapter, only Siaroff ’s (2019) differentiates moderate
b4407_Ch04.indd 70 1/6/2022 9:15:15 AM

1 B
0.9
0.8
0.7
S
R
(S3 + Sr) / (S1 + Sr)
0.6
E
0.5 F
0.4
G
0.3 N Q
0.2
0.1 O P
D
0
A0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1C
(S2 + Sr) / (S1 + Sr)
Figure 4.1. The segmented relative-size triangle with some data points
of significance.
multiparty systems from what the author calls “extreme” multiparty systems,
which is of course important but adequately measured with the effective
number of parties. All constellations with two parties are located along
the x-axis. Constellations with more than two parties can take any point,
depending on the relative sizes of the parties but not on their numbers.
The relative-size triangle (RST) can be easily segmented into equal-
size regions corresponding to the theoretically important types of party
systems. The geometrical center of any triangle is its centroid (G), the point
located at the intersection of its medians. The coordinates of the centroid
are the mean coordinates of the vertices, which makes it the true “average
point” of any triangular graphical representation. While intersecting at the
centroid, the medians divide the triangle into six equal-size segments. All
b4407_Ch04.indd 71 1/6/2022 9:15:15 AM

6"×9"
Table 4.5. The segments of the relative-size triangle defined by the

sizes of three largest components.
Segment of the
relative-size triangle s1 s2 s3
AEG s1 > 0.5 s2 < (s1 + s3)/2 s3 > (s2 − sr)/2
ADG s1 > 0.5 s2 ≤ (s1 + s3)/2 s3 ≤ (s2 − sr)/2
CDG s1 ≥ 0.5 s2 > (s1 + s3)/2 s3 < (s2 − sr)/2
CFG s1 < 0.5 s2 > (s1 + s3)/2 s3 < (s2 − sr)/2
BFG s1 < 0.5 s2 ≥ (s1 + s3)/2 s3 ≥ (s2 − sr)/2
BEG s1 ≤ 0.5 s2 < (s1 + s3)/2 s3 > (s2 − sr)/2
data points lying below the CE line represent those constellations in which
there is a majority party, the points above it, those in which there is not,
and the points on it, those where the majority party takes exactly half of the
seats. The definitions of the six segments of the RST that stem from similar
algebraic transformations are reported in Table 4.5.
While a standard procedure in the science of comparative politics is
to replace the proper names of the phenomena with formal concepts, the
deductive method sometimes makes it necessary to do the opposite. The
relationships between some of the segments of the diagram and the major
types identified in traditional qualitative typologies are not problematic.
They are defined by the vertices of the triangle. The A vertex represents
the constellation in which all seats are taken by one party, which is perfect
one-party dominance. The C vertex is the point of perfect bipartism because
here, there are only two equal-size parties. At the vertex B, we find perfect
multipartism, with constellations of more than two equal-size parties.
The substantive definitions of the triangular segments of the relative-size
triangle, as listed in Table 4.5, are thus. The AEG and ADG segments repre-
sent predominant party systems of two varieties: in the former, the opposition
is fragmented, while in the latter, only a small number of opposition parties
can be observed. In democratic contexts, predominant party systems are
rare and often unsustainable, so that the utility of drawing this distinction
b4407_Ch04.indd 72 1/6/2022 9:15:15 AM

is limited and can be interpreted primarily as an indication of the direction

in which the predominant party system is likely to develop upon achieving
a more balanced relationship among the political forces. At the same time,
predominant party systems are quite likely to be found under authoritari-
anism, and the distinction between the two categories, further in this study
dubbed “predominant party systems with fragmented opposition” (AEG)
and “predominant party systems with concentrated opposition” (ADG), is
important for our understanding of the models of party dominance under
authoritarianism.
The CDG and CFG segments represent what is labeled, both tradi-
tionally and in this study, “two-party systems” and “two-and-a-half ” party
systems, respectively, even though it has to be recognized that earlier research
tended to reduce the latter category to a more limited number of possible
cases. The BFG and BEG segments represent two varieties of multipartism
registered in some of the previous classifications, including those of Rokkan
(1970) and Siaroff (2019). These varieties will be referred to as “balanced
multiparty systems” (BFG), indicating the fact that none of the parties enjoys
significant leverage over others, and “imbalanced multiparty systems” (BEG),
where such a leading party is in place.
These formal definitions, of course, do not liberate me from the necessity
to show “how it works.” While the analysis of the real-life data will follow,
some hypothetical constellations can be illuminating. Rather than selecting
them arbitrarily, I use for this purpose the centroids of the triangular subtype
segments. The relative-size triangle is built in a way that allows each data
point on the diagram to represent several possible constellations. That is
why Table 4.6 reports two constellations for each of the centroid data points.
The selected constellations equal to or closely approximate the theoretical
limits of minor party presence at any given point. The letter designators of
the data points are from Figure 4.1. For each of the constellations, I report
the actual number of parties and two versions of the effective number of
parties, all on the assumption that minor parties starting with s4 are of the
same size as s3, or take residual seats.
b4407_Ch04.indd 73 1/6/2022 9:15:15 AM

6"×9"
Table 4.6. 12 hypothetical party constellations at the segment

centroids of the relative-size triangle diagram.
Party system subtype/the Actual Laakso–
relative-size triangle data number of Taagepera Golosov
point, Figure 4.1 s1 s2 s3 sr parties index index
Predominant party 0.60 0.23 0.17 0.00 3 2.26 1.77
systems with fragmented
0.64 0.13 0.03 0.20 9 2.31 1.77
opposition/N
Predominant party 0.67 0.26 0.07 0.00 3 1.93 1.55
systems with concentrated
0.68 0.24 0.03 0.05 5 1.91 1.53
opposition/O
Two-party system/P 0.55 0.39 0.06 0.00 3 2.19 1.91
0.55 0.38 0.02 0.05 6 2.22 1.92
Two-and-a-half party 0.46 0.41 0.13 0.00 3 2.51 2.30
system/Q
0.45 0.39 0.05 0.10 4 2.68 2.43
Balanced multiparty 0.40 0.36 0.24 0.00 3 2.89 2.62
system/R
0.32 0.24 0.04 0.40 13 5.63 4.97
Imbalanced multiparty 0.43 0.31 0.26 0.00 3 2.87 2.47
system/S
0.36 0.12 0.02 0.50 24 6.49 4.99
The relative-size triangle representations of a large set of democratic

party systems, built in accordance with a number of theoretically sub-
stantiated formal criteria (Golosov, 2013c) and covering the period of
1792–2009, have been provided elsewhere (Golosov, 2014e). Related to
the narrow purpose of this study, Figure 4.2 (Plate 1) graphically rep-
resents the distribution of 148 party constellations of 74 democracies,
selected as explained in the previous section of this chapter. The diagram
demonstrates that while none of the party system formats is entirely alien
to democracies, predominant party systems are not very widespread. The
vast majority of data points on the diagram are located in the segments
representing two-party systems and the two varieties of multipartism. It
can be noticed that the region of balanced multiparty systems is particu-
larly densely populated by democratic party systems, while two-and-a-half
party systems are relatively rare.
b4407_Ch04.indd 74 1/6/2022 9:15:15 AM

1 1
0.9
Plate 1 0.9
Plate 2
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.2 0.4 0.6 0.8 1
1
Plate 3
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.2. The graphic representation of party constellations in democ-

racies and authoritarian regimes. Plate 1: 148 democratic elections,
1987–2014. Plate 2: 294 authoritarian elections, 1945–2019. Plate 3:
57 authoritarian party systems, 1945–2019.
Figure 4.2 (Plates 2 and 3) represents 294 authoritarian party constella-

tions at the level of individual elections and 57 party systems, respectively.
The representations of party systems have been built by averaging the shares
of seats gained by individual political parties throughout the periods under
observation and applying the formulas for the coordinates, as provided
above, to the resulting numbers. This allows for discounting deviant cases
within each of the country-level categories. As follows from the graphs,
authoritarianism generates predominant party systems in a vast majority
of cases. They are rather evenly distributed between the types with and
without fragmented oppositions. Two-party and imbalanced multiparty
constellations are not unusual, but this does not translate to the level of party
systems, which indicates that within most of the country-level categories
b4407_Ch04.indd 75 1/6/2022 9:15:16 AM

6"×9"
such constellations are normally preceded and/or followed by a greater

number of predominant-party constellations. Two-and-a-half and balanced
multiparty constellations are few, and they converge into party systems in
an extremely small number of cases.
Party System Volatility

Party system volatility — a concept that refers to the total change in the per-
centage of seats or votes won or lost by all parties between elections — plays
a central role in empirical research on political parties and elections (Lipset
and Rokkan, 1967; Rose and Urwin, 1970; Bartolini and Mair, 1990). With
application to democracies, the reasons for the importance of party system
volatility and its opposite, stability, are discussed at length by Mainwaring
and Zoco (2007) who argue that high levels of party system stability are ben-
eficial for democracy. Of course, it is also clear that in democracies certain
degrees of electoral uncertainty are necessary to ensure the responsiveness
and accountability of the elites. The very possibility of vote switching is
essential for placing effective checks on parties and their representatives
(Lijphart, 1999). It has to be noticed that the pattern of remarkable stability
of democratic party systems, characterized by Lipset and Rokkan (1967) as
“freezing,” is no longer observable. Starting with the 1970s, electoral vola-
tility has been continuously on the rise in the well-established democracies
(Drummond, 2006), and it is high in many new democracies (Birch, 2003;
Mainwaring and Scully, 1995; Sikk, 2005; Tóka, 1998).
In the studies of electoral volatility, authoritarian regimes were normally
excluded from the samples at the stage of case selection. There is only one
region — Africa — where, as a result of such regimes’ particularly widespread
presence, they have never been ignored in research on electoral volatility
(Mozaffar and Scarritt, 2005; Lindberg, 2007). Contrary to the self-evident
idea that the results of authoritarian elections, related to the lack of alteration
in power, should be stable over time, this stream of research has revealed
relatively high levels of volatility. Moreover, it has been demonstrated that
b4407_Ch04.indd 76 1/6/2022 9:15:16 AM

in Africa volatility is actually higher in authoritarian regimes than it is in

nascent democracies (Bogaards, 2008). The reasons for this tendency will be
discussed in the subsequent chapters of this book. At this point, it is essential
to establish the levels of authoritarian party system volatility empirically, on
the basis of evidence from a wider set of cases.
In contrast to fragmentation and format, the measurement of
party system volatility does not pose a methodological problem. The
standard measure of electoral volatility is the index developed by
Pedersen (1979):
1 x
V= ∑ | v − v |,
2 1 i ,t1 i ,t2
(4.11)

where vi ,t is the percentage vote share of the ith party at the first election, and
1
vi ,t is the percentage vote share of the ith party at the second election. This
2
simple measure belongs to a wider family of indices developed for express-
ing the concept of temporal or structural unevenness, or, speaking in more
concrete and operational terms, “discordance between two sets of figures
that conceivably can be equal” (Taagepera and Grofman, 2003, p. 661). The
properties of some of the phenomena subject to this measurement are such
that the build-up of the respective indices has to involve weighting schemes
in order to emphasize the contribution of the largest components, which is
the case with the index of disproportionality developed by Gallagher (1991).
However, there is no theoretical reason to assume that such weighting is
necessary in the measurement of party system volatility.
In the recent literature, there were proposals to disaggregate this
simple measure into two components, voter volatility and party volatility,
which isolates the question of volatility from that of party replacement (Birch,
2003). Such a strategy, however fruitful for solving certain research tasks
(Powell and Tucker, 2014), is scarcely consistent with the aims of this inquiry,
for the concept of party system volatility embraces both cross-temporal
unevenness in the support of continuous parties and party replacement.
When tracing continuities between political parties in the contexts that
b4407_Ch04.indd 77 1/6/2022 9:15:16 AM

6"×9"
involved renaming and/or coalition formation, I followed several method-

ological desiderata proposed in the related literature (Sikk, 2005).
Table 4.7 shows the aggregate values of the effective number of parties
for the same categories of party systems as in Table 4.2. Tables 4.8 and 4.9
show the results of statistical analysis. The datasets used for the calculation
of values presented in the two tables are identical, but note that the overall
numbers of observations are different because the earliest elections in each
of the included sequences naturally did not yield any relevant numbers.
As previously, and for the same reasons as previously, all calculations have
been performed on seat distributions rather than vote distributions. In
fact, the observed dynamics of the party composition of the assembly is
more directly related to the notion of party system volatility than the cross-
temporal variability of voter preferences (Mair, 1989). At the same time, it
is clear enough that the two phenomena are fundamentally related to each
Table 4.7. Volatility of democratic and authoritarian party systems.

Standard
Democratic party 18.17 16.58 5.00 36.36 9.36 48
systems 1
Democratic party 26.63 22.13 6.94 83.33 17.80 26
systems 2
Authoritarian party 15.21 12.55 0.00 64.99 13.80 237
systems by election
Authoritarian party 17.29 17.55 0.00 42.17 10.41 57
systems by system
Table 4.8. Summary of ANOVA comparison of party systems by volatility.

Sum of Degrees of Mean
squares freedom square F Significance
Between groups 1685.47 2 842.74 5.96 0.00
Within groups 18,104.80 128 141.44 — —
Total 19,790.27 130 — — —
b4407_Ch04.indd 78 1/6/2022 9:15:16 AM


authoritarian party systems by volatility.
Mean Standard
Democratic Democratic party −8.46 2.90 0.01
party systems 1 systems 2
Authoritarian party 0.89 2.33 1.00
systems
Democratic Democratic party 8.46 2.90 0.01
party systems 2 systems 1
systems
Authoritarian Democratic party −0.89 2.33 1.00
party systems systems 1
Democratic party −9.34 2.81 0.00
systems 2
other, even though the relationship is intermediated by electoral system

effects. Many influential studies do not discriminate between them when
performing empirical analysis (Mainwaring, 1998).
As follows from the table, the levels of volatility of authoritarian party
systems are not significantly lower than in the countries at the apex of
democratic development, but decisively lower than in new democracies.
The latter finding is hardly puzzling. High volatility is only natural for the
party systems of less-than-perfect democracies, and particularly for the
party systems of newly democratized countries with their combinations
of nascent electoral politics and the lack of political constraints that could
suppress the entry of new competitive actors. But in authoritarian regimes,
such constraints are sometimes severe, while incentives for sustaining
stable patterns of dominance are strong, which explains the presence of
several cases of zero volatility not only at the level of individual elections
but also, remarkably enough, among authoritarian party systems for
the whole time of existence. Yet the average values of volatility are high
enough, and the presence of extremely volatile systems is apparent from
b4407_Ch04.indd 79 1/6/2022 9:15:16 AM

6"×9"
the data. This is an empirical puzzle to be addressed in the subsequent

chapters.
Party System Nationalization

Party systems vary in the degree to which there is an equal distribution
of party votes across territorial units. Major concepts that enable political
scientists to capture this variation are party system nationalization and its
opposite, regionalization. The mainstream definition of party system nation-
alization is quite straightforward: a party system is fully nationalized if the
shares of the votes received by each of the political parties do not vary from
one territorial unit of the country to another, and it completely lacks nation-
alization if each party receives electoral support in only one of many units.
There are varying degrees of nationalization in between these extremes. For
a long time, research on party system nationalization was focused primarily
on the formation of national electorates, with nationalization processes
representing “a broad historical evolution toward the formation of national
electorates and party systems” leading to the progressive disappearance of
“peripheral and regional specificities” (Caramani, 2004, p. 1).
This stream of research, initiated by the studies of party system nation-
alization in the United States (Stokes, 1967; Vertz et al., 1987), received
a continuation in several pioneering comparative inquiries published in
the 1970s–1990s (Rose and Urwin, 1975; Ersson et al., 1985; Hearl et al.,
1996), yet the decisive breakthrough in comparative research on party
system nationalization occurred relatively recently, after the almost simul-
taneous publication of major studies focused on Latin America (Jones and
Mainwaring, 2003), Western Europe (Caramani, 2004), and the world’s
largest federations (Chhibber and Kollman, 2004).
Since then, the somewhat teleological vision of earlier research, with
party system nationalization being viewed as a crucial indicator of the
maturity of parties and party systems, has been complemented, or even
challenged, by a new approach that conceives the territorial heterogeneity
b4407_Ch04.indd 80 1/6/2022 9:15:16 AM

of party support as a property that can be sustained in party systems, irre-

spective of the levels of their maturity, by a combination of structural factors
(Morgenstern and Swindle, 2005; Chhibber and Murali, 2006; Brancati,
2008a; Lublin, 2017; Schakel, 2018). There is a new and promising stream
of research on federalized/decentralized (Gibson and Suarez-Cao, 2010;
Harbers, 2010) and multilevel (Schakel, 2013) party systems. Many of
the recent studies are focused on the institutional factors of party system
nationalization (Simón and Guinjoan, 2018; Riera, 2018; Su, 2018) and
their interaction with societal characteristics (Golosov, 2016a). Little cross-
national research, however, has been done on party system nationalization
under authoritarianism (see, however, Golosov, 2016d).
In comparison to the measurement of the party system parameters
previously discussed in this chapter, the measurement of party system
nationalization is still in its infancy, as testified by the intensity of the ongoing
debate on the matter (Bochsler, 2010) and the abundance of vastly different
measures on offer. Some of these measures, even if sophisticated and success-
fully applied in previous research, are inconsistent with the purposes of this
analysis. Primarily, this applies to the family of measures, jointly referred to
as “inflation scores,” derived from the empirical fact that sub-national party
constellations are normally less fragmented than national party systems
(Cox, 1997; Chhibber and Kollman, 1998; Moenius and Kasuya, 2004; Kasuya
and Moenius, 2008). These measures quantify the resulting discrepancies by
using the effective number of parties. By design, the analysis performed in this
book conceives sub-national fragmentation as a factor distinct from the effective
number of parties at the national level, which makes it necessary to analytically
separate them from each other. A different approach to the measurement of party
system nationalization (Golosov and Ponarin, 1999; Jones and Mainwaring,
2003) is based on weighting the nationalization scores of individual parties by
their national electoral returns, followed by summing up the products:
x
N = ∑ (N i ′ × si ), (4.12)
1
b4407_Ch04.indd 81 1/6/2022 9:15:16 AM

6"×9"
where N is party system nationalization score, and Ni′ is the nationalization

score of the ith party. Obviously, this formula assumes the decomposability
of party system nationalization scores into individual-party components.
The literature offers quite a substantial amount of individual-party
nationalization indices. However, some of them (Rose and Urwin, 1975;
Ersson et al., 1985) cannot be built into the above formula for the reason
that they are not normalized, i.e., their values do not run from 0 to 1. The
requirement of normalization is fulfilled by relatively few indices of individ-
ual-party nationalization/regionalization. The party system nationalization
score, proposed by Jones and Mainwaring (2003) and defined as the inverse
of the Gini index of inequality, can meet this principle, yet it has not been
properly normalized in the original publication. The algebraic build-up of
the Gini coefficient is such that its upper bound, rather than being set at
1 for all constellations with maximum inequality, has an upper limit of
(1 − 1/n). In order to normalize the Gini coefficient, it is necessary to subject
the values of the Gini coefficient to simple additional transformation, which
is achieved by multiplying them by n(n − 1) (Deltas, 2003). This, upon some
algebraic transformations, yields the following formula for the normalized
Gini coefficient (NG):
n
2∑ (si × i)
n +1
NG = n
1
− , (4.13)
n −1
∑ si (n − 1)
1
where n is the number of units, and i stands for the rank from 1 for the
smallest component to n for the largest one. Correspondingly, the definition
of the normalized party nationalization score for the ith party is
NPNSi = 1 − NGi . (4.14)

There are no substantive obstacles for the use of this index as a substi-
tute for Ni′. At the same time, there is an available alternative that fulfills
all requirements to the aggregate quantities of this kind and produces
b4407_Ch04.indd 82 1/6/2022 9:15:16 AM

remarkably similar sets of values on real-life datasets. The index of party

nationalization (Golosov, 2016b) employs the same index-building principle
as the Herfindahl–Hirschman index of concentration discussed in the first
section of this chapter. It is defined as
 n  2 n 
n −   ∑ si  ∑ si2 
 1  1
IPN = 1 − , (4.15)
n −1
which yields the following index of party system nationalization:
  n 2 n  
n
  n −   ∑ si  ∑ si2   
 
IPSN = ∑  1 −
1 1  p , (4.16)
1 
 n −1  i 

where sigma stands for summation; n stands for the number of electoral
districts or other territorial units; si stands for the percentage or fractional
share of the vote received by the ith party in each of the territorial units; and
pi, for the fractional share of the vote received by the ith party nationally.
The values of the index run from zero for those systems that completely
lack nationalization, i.e., each of the parties receives support in only one
of many territorial units, to one for fully nationalized systems in which
all parties enjoy equal support across all territorial units. From the point
of view of computational convenience, the Gini coefficient is as highly
labor-consuming as any index that involves ranking the units of analysis by
size, which is a procedure that has to be performed for each of the parties
separately. The calculation of the index of party system nationalization does
not require the reordering of the data. Thus, for practical reasons, it can be
preferable to the normalized party nationalization score, even though from
the substantive point of view, the two indices are largely equivalent and can
be used interchangeably.
The data demands of the index of party system nationalization are
different from and much higher than those of the indices discussed in the
previous sections of this chapter. Firstly, the index cannot be calculated on
b4407_Ch04.indd 83 1/6/2022 9:15:16 AM

6"×9"
the basis of seat distributions. Only the shares of the votes cast for individual
parties can be employed. Secondly, and more importantly, the calculation
of the index of party system nationalization necessarily involves not only
the shares of the votes obtained by parties at the national level, but also the
distributions of the votes in each of the territorial units, however defined.
For a long time, the unavailability of sub-national electoral statistics practi-
cally restricted the study of party system nationalization to a small number
of well-established democracies, and even for them, only relatively recent
information was available, even though the effort of Caramani (2000) helped
to improve this situation. Fortunately, in the recent years some remarkable
advances in the development of publicly available election statistics datasets
have occurred.
In this book, by far the most extensive use has been made of the
Constituency-Level Elections Archive, CLEA (Kollman et al., 2019). Some of
the data have been derived from the European Election Database (http://o.nsd.
no/european_election_database/about/; accessed at different times in 2011–
2018). These data are collected from original sources, prepared and made
available by the Norwegian Centre for Research Data that is not responsible
for the analyses/interpretation of the data presented here. Similar disclaimers
can be applied to the data from the African Election Database (http://african
elections.tripod.com/about.html; accessed at different times in 2011–2018)
and Electoral Geography 2.0 website (http://www.electoralgeography.com/
new/en/; accessed at different times in 2011–2021). To a significant extent,
the data have been collected from national election statistics sources.
For as many as 15 authoritarian party systems, the index of national-
ization could not be obtained due to the lack of the relevant data on any
of the elections held during the periods of these systems’ existence. While
some of the remaining authoritarian party systems offered much higher
levels of data availability, for many others, the sub-national data were avail-
able for no more than one election. In order to secure a balanced structure
of the dataset, and, acting on an informed assumption that party system
nationalization is a very stable parameter, I created a dataset that includes
b4407_Ch04.indd 84 1/6/2022 9:15:16 AM

single observations from each of the party systems. Whenever more than
one observation was available, I tended to select an observation from the
median election within the party system’s lifespan. All values have been
obtained from the same set of country/year cases as in the previous analysis,
but there are two exceptions.
For Gabon, I used the data from the 2018 elections that were held in the
conditions of relative and reversible democratization and thus did not meet
the criteria for the inclusion into the main dataset. However, my comparison
of the structure of party alternatives and other parameters of importance
in the 2018 Gabonese elections with those of the previous elections held in
authoritarian conditions has revealed no major difference whatsoever, which
made it permissible to use the 2018 data as a substitute. For Guatemala,
I found it possible to use the sub-national results of the 1966 presidential
elections because they were contested by the same parties as the concurrently
held legislative elections.
Two technical comments related to the mechanics of electoral systems
are in order. First, those mixed electoral systems in which each of the
voters cast two votes, which is normally but not necessarily the case, produce
different vote distributions in the proportional representation and majori-
tarian sections of elections. Depending on data availability, I dealt with this
problem in two ways. For all democracies and some authoritarian regimes,
the averages for both sections are used. For several other authoritarian
regimes, the indices of nationalization have been computed on the basis of
the available data from either of the sections. Second, in most cases I used
electoral districts as units for the calculation of the indices of nationalization,
but in some other cases, particularly when dealing with those countries that
employ proportional representation systems in single nationwide districts,
the units of analysis were defined on the basis of the existing administrative
territorial structures.
Table 4.10 reports the aggregate values of the index of party system
nationalization for democratic and authoritarian party systems. Tables 4.11
and 4.12 show the results of related statistical analysis. As follows from the
b4407_Ch04.indd 85 1/6/2022 9:15:16 AM

6"×9"
Table 4.10. Nationalization of democratic and authoritarian party

systems (index of party system nationalization).
Standard
Table 4.11. Summary of ANOVA comparison of party systems by

nationalization.
Between groups 0.41 2 0.21 8.97 0.00
Within groups 2.59 113 0.02 — —
Total 3.00 115 — — —

authoritarian party systems by nationalization.
Mean Standard
Democratic party Democratic party 0.09 0.04 0.04
systems 1 systems 2
systems
Democratic party Democratic party −0.09 0.04 0.04
systems 2 systems 1
systems
Authoritarian party Democratic party −0.13 0.03 0.00
systems systems 1
Democratic party −0.04 0.04 0.90
systems 2
b4407_Ch04.indd 86 1/6/2022 9:15:17 AM

table, and consistent with the theoretical claims of the early literature in the
field, the level of democracy is positively associated with the level of party
system nationalization. But this applies primarily to the well-established
democracies belonging to the upper category on the Freedom House scale.
Less-than-perfect democracies in the second category also outperform
authoritarian regimes on the parameter of party system nationalization, but
the difference is not strong. Moreover, the data show that on this parameter,
the diversity within the set of authoritarian party systems is greater than
it is within the subsets of democracies, which implies that under authoritar-
ianism, the structure of incentives for the nationalization of party politics
is quite complex.
b4407_Ch04.indd 87 1/6/2022 9:15:17 AM

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Chapter 5
Political Regimes and

T
his chapter aims at establishing associations between political
regimes and the properties of party systems in a way that allows
for causal inference. To this end, it is essential to reconsider the
notion of political regimes in accordance with the research tasks at hand.
While the very existence of different authoritarian political regimes is the-
oretically unquestionable and methodologically important (Gleditsch and
Ward, 1997), making empirical distinctions among them can be problematic
because of the idiosyncratic, poorly articulated, or transient properties of
some individual cases. For certain research purposes, it is quite possible to
complement well-defined types with categories representing their “anoma-
lous mixtures” (Wilson and Piazza, 2013, p. 947). But for statistical analysis
in which authoritarian regime types are employed as major explanatory
variables, it is essential to build these variables on the basis of mutually
exclusive categories (Collier, 2008), which makes it imperative to assign
each observation to a single type.
Following the above methodological desiderata, the first section of this
chapter refines the conceptual framework and then applies classificatory
procedures to the observed cases. Then I lay out my theoretical expectations
regarding the political regime determinants of authoritarian party systems.
In the concluding section of this chapter, I test these expectations by per-
forming statistical analyses. As a technical note, it should be mentioned
that when discussing individual country cases, I refer to the time periods
when the respective party systems were in existence, as listed in Table 2.1.
89
b4407_Ch05.indd 89 1/6/2022 9:15:29 AM

6"×9"
For instance, the label Iran refers not to the contemporary Islamic Republic,
excluded from this study because of the lack of party-structured elections,
but rather to the period of 1956–1971 when the monarchical regime exper-
imented with a two-party system.
Authoritarian Political Regimes and

Regime Genomes
As explained earlier in this book, I start with a simple four-component clas-
sification of authoritarian political regimes: monarchies, military regimes,
single-party regimes, and personalistic regimes. Within this order of cate-
gories, the notion of electoral authoritarianism in the narrow sense refers to
personal dictatorships that cannot be ascribed to other categories because of
the lack of the mechanisms of power acquisition/succession that are specific
to these categories, i.e., hereditary succession, military takeovers, or intra-
party arrangements. Given that elections play a particularly important role
in the functioning of contemporary personalistic regimes, and related to the
tasks of this study, I follow Hadenius and Teorell (2007) in dividing person-
alistic regimes into two categories, party electoral authoritarian regimes and
non-party electoral authoritarian regimes. Monarchies, military regimes, and
single-party regimes can also conduct elections, so that the broad notion of
electoral authoritarianism is applicable to them on the condition that they
do permit limited inter-party competition.
The basis for the above classification is laid by the origins of politi-
cal regimes, and this provides us with the easiest way to draw empirical
distinctions among them, as practiced in a number of recent studies and
related datasets (Geddes, 1999; Hadenius and Teorell, 2007; Magaloni and
Kricheli, 2010). I consider this approach as a good example of how theoretical
parsimony can lead toward sound empirical solutions. Is there a danger of
oversimplification? For illustration, consider Indonesia under the rule of
Suharto. This regime obviously originated from a military takeover, thus
falling into the category of military regimes. Throughout its long lifespan, the
b4407_Ch05.indd 90 1/6/2022 9:15:29 AM

Political Regimes and Authoritarian Party Systems 91
regime of Suharto continued to rely on its originally very cozy relationship

with the armed forces, which was emphasized by the fact that within the
dominant pro-regime party, Golkar, the top military officers were always at
the core positions of influence (Jenkins, 1984). At the same time, Suharto’s
personal control over the Golkar increased over time, often at the expense
of the military. This was accompanied by a within-party split along military
and civilian lines (Suryadinata, 1997). Does it mean that the regime changed
its nature, so that it has to be considered as an electoral authoritarian regime
in the narrow sense, at least in the late phase of its existence?
My answer to this question is negative. Despite the gradual weakening of
the ties between Suharto and the military, they remained the only force that
could practically remove the dictator from power, which actually happened
in 1998. The student protests triggered the collapse of the regime, but it was
due to the fact that some of the top military commanders were consistently
conciliatory with the student protestors and ultimately abandoned their
loyalty to Suharto that he lost his power (Lee, 2009). Thus, the origin of the
regime defined both its mode of operation and its end point, but the same
could be true with regards to its possible survival: were the military to retain
their loyalty, Suharto would have remained in power.
To extend the argument, monarchies survive if hereditary succession
remains operational; military regimes survive as long as they are able to
transfer power within the military corporation; single-party regimes survive
by employing intra-party mechanisms of power transition; and personal
dictatorships survive as long as the narrow circle of the ruler’s trusted
officials and friends are able to make decisions on power distribution and
succession. The logic of decision-making in non-monarchical regimes has
been formally modeled in an influential study of Acemoglu et al. (2008).
From the perspective of this study, it is important to emphasize that the
survival of authoritarian regimes is not conditioned by election outcomes.
Thus, the first step in ascribing real-life political regimes to different
categories, consistent with the approach that is prevalent in the literature,
is to look at their origins. The types of authoritarian regimes are contingent
b4407_Ch05.indd 91 1/6/2022 9:15:29 AM

6"×9"
on their origins not only in the sense of mere determination, but also in
the wider sense that all autocracies display continuous patterns of power
acquisition and maintenance, and these are the same as at the time when
they first come into existence. From the empirical point of view, however,
this approach causes two difficulties. First, governments that have had
democratic origins may end democracy by abolishing or severely limiting
democratic procedures (Huntington, 1993, p. 8), which is one of the ways in
which personal dictatorships come into existence. This logically implies that
democracy has to be identified as one of the authoritarian regime clusters,
which obviously looks like an oxymoron.
This leads us to the second difficulty. It has to be kept in mind that all
classifications in political science are instrumental, and their instrumen-
tality depends on how they correspond to particular research goals (Finer,
1983). Since this study is focused on authoritarian party systems, not on the
regimes per se, it is essential to juxtapose the two phenomena in a way that
allows for establishing direct associations. Authoritarian regimes do not
necessarily involve formally competitive elections, so that each of them —
with the exception of electoral authoritarianism in the narrow sense — can
lack a party system whatsoever or, which is more important for this study,
can develop a party system at some later stage of development. The moment
of determination in which regime characteristics exert their immediate
impact upon party systems’ properties is not necessarily coincidental with
the emergence of the regime as such.
Taking this into consideration does not imply a drastic change of per-
spective. Going back to the previous example, Indonesia under Suharto
would still be categorized as a military regime because during the period
immediately preceding the earliest elections, from 1965 to 1970, the regime
presented a clear-cut instance of a military dictatorship, and it was in this
capacity that the influence of regime characteristics upon the emerging
party system was exerted.
There are less straightforward cases, though. Consider the Republic
of Congo (Brazzaville). The regime of Marien Ngouabi emerged as a result
b4407_Ch05.indd 92 1/6/2022 9:15:29 AM

of a military coup. The two subsequent successors of Ngouabi, including

Denis Sassou Nguesso, both came to power relying predominantly on their
control over the military (Decalo, 1985a). Since, however, the ruling mil-
itary claimed their ideological allegiance to communism in its Soviet ver-
sion, they carefully emulated the institutional structures of the communist
regime, including the “guiding role” of the Congolese Party of Labor and the
communist-style elections without choice (Morrison et al., 1989). Similarly
to many other African countries, Congo introduced multiparty elections in
1990. If conducted in non-democratic conditions, these elections would have
been influenced by the previous communist-style practices, thus assigning
Congo to the category of single-party regimes.
In fact, however, the 1990 elections were relatively free and produced a
short-lived democracy that collapsed after Sassou Nguesso won in a brutal
civil war and reassumed power. The Congolese Party of Labor did not play
any significant role as a military force in this conflict (Clark, 1998). It was
only after Sassou Nguesso’s military takeover that the Congolese Party of
Labor, now devoid of its previous ideological commitments and largely
serving as a personal political vehicle of Sassou Nguesso, was restored as
a dominant party within the new context of electoral authoritarianism. In
this case, therefore, it stands to reason to identify the Congolese regime, in
its relationship to the origins of the respective party system, as a military
regime rather than a single-party regime.
The example of Congo shows not only that in some cases assigning
party systems to political regime categories requires some penetration into
the complexities of national political history, but also, and more importantly,
that the very notion of political regime has to be refined in order to fulfill
the research tasks at hand. In order to resolve these difficulties and to avoid
terminological confusion, I supplement the terms “regime type” and “regime
origin” with the term “regime genome.” The term “genome,” derived from
biology, is selected by me for its substantive suitability. In biology, the genome
is modeled as an ordered sequence of genes that forms a material substrate of
heredity in any living organism. In this study, the notion of regime genome
b4407_Ch05.indd 93 1/6/2022 9:15:29 AM

6"×9"
refers to the causal conjunction of the substantive characteristics of the

regime’s origins and the related characteristics of the regime’s functioning,
which does not include the situational/accidental traits. To put it briefly,
political regime genome is an integral characteristic of the regime’s origin
and mode of operation.
Note that even upon terminological refinement, I find it useful to use
the term “post-democracy” when referring to those authoritarian party sys-
tems that directly inherit the characteristics of preceding democratic party
systems. But the prefix “post” could be also applied to the labels of all other
regime genomes with the exception of monarchy. This is avoided simply for
the sake of a smoother narration. There are six regime genomes to be dealt
with in the further analysis: monarchy, military regime, single-party regime,
post-democracy, party electoral authoritarian regime, and non-party elec-
toral authoritarian regime. Taking into account that the last three categories
represent major varieties of contemporary personalist dictatorships, this
taxonomy corresponds to the rather conventional classification of authori-
tarian political regimes presented in Chapter 2 of this book.
Assigning Party Systems to Regime

Genome Categories
In this study, the category of monarchies comprises all cases where, on the
one hand, significant powers are concentrated in the hands of the mon-
arch, and, on the other hand, multiparty elections are in place. Within
the set of authoritarian party systems analyzed in this study, there are
only four that fully satisfy these conditions: Iran, Morocco, Nepal, and
Thailand. Three other countries — Cambodia, Malaysia, and Sierra Leone in
1967–1971 — were monarchies by institutional design during the periods
observed in this study, but their monarchs lacked significant formal and/or
effective powers. These cases are therefore categorized as different political
regimes. Thailand may be problematic because of the continuous involve-
ment of the military in politics. This has been manifest in several military
b4407_Ch05.indd 94 1/6/2022 9:15:29 AM

coups, including one that occurred during the observed period. However,
in-depth studies demonstrate quite convincingly that even when the mili-
tary assumed political leadership, the monarch and his court continued to
play a pivotal role in the politics of Thailand and effectively prevailed over
the military in a system referred to as “royal hegemony” (Tejapira, 2016).
Authoritarian party systems can be assigned to the category of military
regimes if they emerge in the conditions when, at the time of the earliest
elections during the observed period, the armed forces either directly
control the national executive or exert decisive impact over the political
decision-making process even if formally conceding power to a civilian
administration. While the first element of this definition is self-evident, the
second may seem problematic from the empirical standpoint. However, the
approach according to which the decisive impact exerted by the military
upon formally civilian administrations constitutes a variety of military
rule is well substantiated theoretically and developed empirically in the
literature on democratic transitions (O’Donnell and Schmitter, 1986; Higley
and Gunther, 1992). It is quite clear, for instance, that the party systems
of El Salvador and Guatemala, in which the most influential parties often
originated from military coups and/or were headed by these coups’ leaders,
did not change their nature due to the mere fact that these leaders kept
winning carefully staged presidential elections (Ruhl, 2004). The category
of military regimes includes the authoritarian party systems of Algeria,
Brazil, Congo (Brazzaville), Dominican Republic, El Salvador, Gambia,
Guatemala, Indonesia, Mauritania, Nicaragua, South Korea, Sudan, and
Yemen.
Single-party regimes as determinants of authoritarian party systems can
be registered on the condition that these systems emerged after periods when
elections were held on a non-competitive single-party basis. Throughout
the period under observation, single-party regimes emerged in several
ways. Only few of them were spanned by the two most salient varieties of
party-based authoritarianism of the 20th century, communist regimes and
right-wing dictatorships influenced by the examples of inter-war Germany
b4407_Ch05.indd 95 1/6/2022 9:15:29 AM

6"×9"
and Italy. The latter category is represented only by Portugal. Indeed, the
second European right-wing dictatorship that survived the end of World
War II, Spain, made a transition to democracy without entering a phase of
electoral authoritarianism. Two non-European right-wing regimes rooted in
the political realities of the Cold War period, Paraguay and Taiwan, evolved
by replacing their single-party systems with authoritarian party systems
(Dickson, 1993; Lambert, 2006).
Rapid transitions to democracy were experienced by a vast majority of
post-communist countries of East Central Europe, with Serbia standing as
the only full-fledged exception. In addition, Angola can be characterized as a
regime that, before its transition to electoral authoritarianism, not just emu-
lated a communist regime but effectively functioned as a communist-type
party dictatorship (Ishiyama, 2005). Three other left-wing regimes, those of
Bolivia, Egypt, and the Seychelles, were established by armed takeovers and
had some similarities to communist regimes without sharing most of their
essential characteristics (Mitchell, 1977; Dekmejian, 1968; Bulbeck, 1984).
Several African countries, such as Côte d’Ivoire, Cameroon, Djibouti, Gabon,
and Togo, established single-party regimes without adhering to left-wing
ideologies. In all these regimes, failed (or, rather, insincere) democratiza-
tions of the early 1990s spanned authoritarian party systems (Morse, 2015).
Tunisia, where multiparty politics were introduced earlier, in the 1980s,
typologically belongs to the same category of “façade democracies” that were
established, to quite a significant extent, with the purpose of making these
regimes more attractive for international financial institutions and foreign
investors (Sadiki, 2002).
In five African countries, Burundi, Chad, Ethiopia, Rwanda, and
Uganda, single-party regimes were created by political-military move-
ments that came to power after their victories in civil wars. With different
speeds, all of them evolved in the direction of electoral authoritarianism.
Among these cases, Uganda can be problematic because before introduc-
ing multiparty elections, the regime of Yoweri Museveni and his National
Resistance Movement organized several elections on a formally non-party
b4407_Ch05.indd 96 1/6/2022 9:15:29 AM

basis, which allows for assigning Uganda to a different category. However,

the “movement system,” established by Museveni in an apparent attempt to
create an alternative to the traditional institutions of liberal democracy, was
heavily dominated by the ruling organization during the whole period when
other parties were not allowed to run in elections (Makara et al., 2009). In
practice, this means that Uganda belongs to the category of single-party
regimes in the same way as Cuba where elections are also conducted on a
formally non-party basis.
Several important authoritarian regimes of the past had democratic
origins (Huntington, 1993), and there are theoretical reasons to envision
a scenario in which a political party, upon coming to power by means of
democratic elections, transforms the country’s political regime into an
autocracy but maintains the provision for multiparty elections as long as
they are viewed as no threat to the political monopoly. This study does
indeed identify such cases, but they are remarkably few. Two of them are
post-Soviet countries, Armenia and Russia. In the latter case, the formation
of the dominant pro-regime party was finalized already after the collapse of
the country’s short-lived democracy, but this party was created by merging
two organizations that were launched in democratic conditions, and the
leader of the regime, Vladimir Putin, first came to power in relatively free
elections (Golosov, 2011b). In the third country belonging to this category,
Guyana, the democratic origins of the regime of Forbes Burnham and his
People’s National Congress party are quite apparent (Singh, 2008).
In fact, a large group of authoritarian party systems emerged in
association with those regimes that already conducted authoritarian
elections at the time of these party systems’ inception, i.e., party electoral
authoritarian regimes and non-party electoral authoritarian regimes. In
the former case, the nascent party systems utilized the organizational
and electoral legacies of the previous phases of authoritarian party pol-
itics, while in the latter case, authoritarian party systems emerged from
scratch because during the previous phases, the role of political parties
in elections was negligible.
b4407_Ch05.indd 97 1/6/2022 9:15:29 AM

6"×9"
A majority of party systems assigned to the category of party electoral

authoritarian regimes emerged in the conditions of colonial rule. For a good
and rather self-evident reason, the Varieties of Democracy (V-Dem) project
tends not to classify most of the colonies as democracies. At the same time,
all major colonial powers with the exception of Portugal did endow their
overseas possessions, especially but not necessarily in the phase immediately
preceding independence, with representative institutions and significant
degrees of autonomy, so that many of these countries did hold elections
contested by political parties. By the beginning of the 1960s, party systems
centered around major pro-independence movements were already in place
in the countries of the French Union (Chafer, 2002). However, most of these
systems, with the only exception of Madagascar, were terminated either by
military coups or by transitions to single-party rule.
The British colonial rule was more prolific in this respect. Not
only the dominant parties but also the major opposition parties that
entered the post-independence electoral arenas of Malaysia, Singapore,
and South Africa emerged in the conditions of colonialism (Maryanov,
1967; Woodward, 1980). The main political party of the Philippines, the
Nationalist (Nacionalista) Party, emerged and developed under the auspices
of the United States-controlled Commonwealth (McCoy, 1989). With some
reservations stemming from the idiosyncrasies of the country’s path to
independence, the two-party systems of Zimbabwe can be added to this
category: the first of them, because the dominant role of the Rhodesian
Front and its immediate predecessor, the Dominion Party, was established
well before the self-proclaimed independence of the white minority state
(Patel, 1979); and the second of them, because the demolition of that
state, largely equivalent to the achievement of independence for the black
majority, was presided by the political parties that were formed during the
previous period and competed in the 1980 elections under the auspices
of the temporarily restored British administration (Gregory, 1981). The
major political parties of Sierra Leone, while originating from the colonial
era, passed through a brief period of non-democratic but party-structured
b4407_Ch05.indd 98 1/6/2022 9:15:29 AM

elections before the establishment of the authoritarian regime of Siaka

Stevens (Riley and Parfitt, 1987).
The three remaining party systems in this category, Cambodia,
Colombia, and Mexico, represent very different trajectories of development.
The main political parties of Colombia emerged long before 1945 and occa-
sionally competed in elections, even though the outcomes of power struggles
between them were not decided in electoral arenas (Posada-Carbó, 1997). In
Mexico, opposition parties were allowed to participate in elections already
in the 1930s, which makes it possible to view the post-1945 experience of
the country’s authoritarian party system as an extension of a long period
of party-structured authoritarian electoral politics. The political domina-
tion of the Cambodian People’s Party was well entrenched in its role of the
ruling party under the 1979–1991 pro-Vietnamese communist regime, but
the party system observed in this study emerged only after the multiparty
elections conducted in 1993 by the United Nations Transitional Authority
in Cambodia (Roberts, 1994). This justifies this system’s assignment to the
category of party electoral authoritarian regime.
The category of non-party electoral authoritarian regimes comprises
almost exclusively post-Soviet cases: Azerbaijan, Kazakhstan, Tajikistan, and
Uzbekistan. Despite significant dissimilarities in the political trajectories of
these countries, the patterns of the development of their party systems are
very similar. All of them emerged after clearly authoritarian periods, normally
in the first decade after the collapse of the Soviet Union, when dominant
parties did not form and elections were contested mostly by independents and
won by those of them who supported the incumbent authorities (Sultanova,
2014; Golosov, 2020). The remaining case, South Vietnam, has been assigned
to this category due to the lack of party-structured elections in the course
of the colonial era (Anh, 1985). The resulting classification of authoritarian
party systems by political regime genomes is presented in Table 5.1.
In order to proceed with this analysis, it is important to empirically
clarify whether the above method of identifying political regime genomes
is independent from the regime characteristic that is central to this study,
b4407_Ch05.indd 99 1/6/2022 9:15:29 AM

6"×9"
Table 5.1. The classification of authoritarian party systems by political

regime genomes.
Political regime
genomes Party systems
Monarchy Iran, Morocco, Nepal, Thailand
Military regime Algeria, Brazil, Republic of Congo, Dominican Republic,
El Salvador (1961–1978), El Salvador (1982–1991),
Gambia, Guatemala, Indonesia, Mauritania, Nicaragua,
South Korea, Sudan, Yemen
Single-party regime Angola, Bolivia, Burundi, Cameroon, Chad, Cote d’Ivoire,
Djibouti, Egypt, Equatorial Guinea, Ethiopia, Gabon,
Paraguay, Portugal, Rwanda, Serbia, Seychelles, Taiwan,
Togo, Tunisia, Uganda
Post-democracy Armenia, Guyana, Russia
Party electoral Cambodia, Colombia, Madagascar, Malaysia, Mexico,
authoritarian regime Philippines, Sierra Leone, Singapore, South Africa,
Zimbabwe (Rhodesia), Zimbabwe
Non-party electoral Azerbaijan, Kazakhstan, South Vietnam, Tajikistan,
authoritarian regime Uzbekistan
Table 5.2. Index of electoral democracy by political regime genomes.

Standard
Mean Median Minimum Maximum deviation
Monarchy 0.238 0.250 0.060 0.400 0.108
Military regime 0.226 0.200 0.130 0.360 0.063
Single-party regime 0.238 0.235 0.100 0.390 0.072
Post-democracy 0.325 0.330 0.240 0.390 0.049
Party electoral authoritarian 0.271 0.260 0.150 0.400 0.066
regime
Non-party electoral 0.234 0.230 0.160 0.310 0.042
authoritarian regime
the level of democracy as quantitatively expressed in the V-Dem index of

electoral democracy. The results of the analysis, performed on the basis of
election-level observations, are presented in Tables 5.2–5.4. As follows from
the table, authoritarian party systems originating from democracies are
b4407_Ch05.indd 100 1/6/2022 9:15:29 AM

Table 5.3. Summary of ANOVA comparison of political regime genomes

by the index of electoral democracy.
Between groups 0.16 5 0.03 6.50 0.00
Within groups 1.38 288 0.00 — —
Total 1.53 293 — — —

genomes by the index of electoral democracy.
Mean Standard
Monarchy Military regime 0.01 0.02 1.00
Single-party regime 0.00 0.02 1.00
Post-democracy −0.09 0.03 0.01
Party electoral −0.03 0.02 0.88
Non-party electoral 0.00 0.02 1.00
Military regime Monarchy −0.01 0.02 1.00
Single-party regime −0.01 0.01 1.00
Non-party electoral −0.01 0.02 1.00
Single-party Monarchy 0.00 0.02 1.00
regime Military regime 0.01 0.01 1.00
Post-democracy Monarchy 0.09 0.03 0.01
Military regime 0.10 0.02 0.00
(Continued)
b4407_Ch05.indd 101 1/6/2022 9:15:29 AM

6"×9"
Mean Standard
Party electoral 0.06 0.02 0.20
Party electoral Monarchy 0.03 0.02 0.88
authoritarian Military regime 0.05 0.01 0.00
regime
Non-party elec- Monarchy 0.00 0.02 1.00
toral authoritar- Military regime 0.01 0.02 1.00
ian regime
indeed significantly different from nearly all other systems, with the exception
of those of party electoral authoritarian regimes. These two varieties of regime
genomes are associated with relatively high levels of electoral democracy,
even though the differences between party electoral authoritarian regimes
and two other genomes — monarchy and non-party electoral authoritarian
regime — are not statistically significant. No other significant differences
among the regime genomes can be registered, which indicates that the above
classification is largely independent from the levels of democracy.
Mode of Power Acquisition

Political regimes, and by implication, political regime genomes, are mul-
tifaceted phenomena. This is true both for democracies and autocracies.
b4407_Ch05.indd 102 1/6/2022 9:15:29 AM

Some of the aspects of political regimes may be inconsequential for party

systems. In the following analysis, I focus on those aspects that, as it can be
theoretically expected, do significantly influence party systems in the ways
that affect their most salient, measurable properties — fragmentation, for-
mat, volatility, and nationalization. I do not presume that the associations
between the peculiarities of the authoritarian regime genomes and party
system properties are uniform in the sense that, to give an example, the
regime genome determinants for party system fragmentation are the same
as for nationalization. Of course, the effects can converge, but this is subject
to empirical inquiry. Therefore, in order to reveal the overall trail of causality,
it is essential to identify the relevant aspects of the regime genomes and to
consider these aspects in analytical separation from each other. Of course,
for such an analysis to be productive, such aspects have to be variable across
the regime genomes.
I start with an aspect that, for the purposes of my analysis, can be dubbed
“mode of power acquisition.” According to Merkel (2004, p. 38), democracies
are characterized by the presence of electoral regime, defined as a mode of
power acquisition that makes the access to public power positions of the state
dependent on the results of elections. Under the electoral regime, elections
are decisive in the sense that the access to power is effectively — not only
formally — restricted to those who win free elections and to their political/
administrative appointees. This applies in full to none of the types of autoc-
racy, but their modes of power acquisition display a significant variation on
this parameter. I theoretically expect that this variation can be of relevance
for party system properties.
From this perspective, monarchy stands out as a political regime where
access to the top positions of executive power is even formally independent
from the results of elections. The most illustrative contemporary case is
Morocco where the 1962 Constitution explicitly endows the monarch with
the role of the guardian and arbiter of the country’s system of government
due to genealogical descent from the Prophet Muhammad, so that the ruling
representative of the dynasty is also the caliph, the legitimate successor to
b4407_Ch05.indd 103 1/6/2022 9:15:29 AM

6"×9"
the Prophet’s secular and religious authority (Joffe, 1998). None of the sev-
eral constitutional reforms experienced by Morocco since 1962, including
the most recent reform of 2011, reduced the essential prerogatives of the
monarchy, thus preserving its control of the Moroccan political scene intact
(Maghraoui, 2011). The Constitution of Thailand remained notoriously
unstable for the whole period after it was first adopted in 1932, and even
during the brief period covered in this study, it was reshuffled several times
(Harding and Leyland, 2011), but none of these constitutional changes effec-
tively diminished the role of the monarch and his powerful advisory body,
the Privy Council, in the country’s government (Hewison, 1997).
The cases of Iran and Nepal represent different — and in a way,
contrasting — modes of preserving effective royal powers in the modern
world. The 1906 Constitution of Iran, even after a significant expansion of
royal powers following the 1949 amendments, endowed the Shah with powers
exceeding those of contemporary European monarchs but still providing for
a parliamentary responsible government (Faridul Haq, 1968). However, the
royal coup of 1953 greatly enhanced the effective powers of the monarch
(Etges, 2011), as a result of which the political regime of Iran was often char-
acterized as the royal dictatorship (Amjad, 1989). The 1990 Constitution of
Nepal, epitomizing the strong concessions made by the royal government
under the pressure of pro-democracy movement, provided for mostly cer-
emonial powers of the King. However, the effective powers of the monarch
remained in place due to the inability of Nepal’s political parties to cope with
the mounting problems of the country, including the Maoist insurgency. In
2001, King Gyanendra dissolved the parliament, refused to call fresh elections
and took over the government (Parajulee, 2010). Thus, in all four monarchies,
the effective powers of the monarchy remained strong, and with the exception
of Nepal, this state of affairs was constitutionally enshrined.
Unlike monarchies, military regimes, as long as they conduct elections,
do not refrain from viewing the popular will as the sole source of legitimate
power. Exceptions to this rule, such as in Myanmar with its military enjoying
a constitutional privilege to maintain veto power in parliament (Crouch,
b4407_Ch05.indd 104 1/6/2022 9:15:29 AM

2019), are few. However, the restoration of elections does not necessarily
change the patterns of power acquisition established after military takeovers.
The long-standing dictatorship of Brazil provides the most telling example
of the situation in which national legislative elections formally served as the
sole source of power but effectively led to the parliamentary confirmation
of top military commanders as presidents of the country. In fact, the fate of
the presidency was decided not by the popular will but rather by back-room
deals within the military leadership (Skidmore, 1988). Some of the military
regimes, from Indonesia to Mauritania, did not allow for alternation in the
top positions of power within the ruling military corporation. In such cases,
direct presidential elections were normally in place, so that the military rulers
remained in power under the guise of a formally civilian administration.
It is important to emphasize, however, that military control over the
government need not be direct. Even if formally ceding power to the elected
government, the military can perform as a veto player. In this way, as put by
Croissant et al. (2010, p. 953), “the electoral regime’s function as a safeguard
for public control of and citizens’ participation in politics is undermined.”
For Dahl (1989, p. 250), the failure of civilians to effectively control their
armed forces was sufficient to account for the existence of non-democratic
regimes in many countries. The instances of military involvement in the
politics of Latin America are particularly well documented in the literature
(Fitch, 1998; Loveman, 1994; Smith, 2005). What has to be emphasized here
is that the modes of military rule display a wide variation, ranging from
outright dictatorial control over the executive to institutionalized prerog-
atives that establish military influence in certain political decision-making
matters (Stepan, 1988).
In some of the regimes, different factions of the ruling military found
it expedient to create or control political parties that actually competed
in elections, thus endowing the citizens with the right to decide which of
them would enjoy priority. For example, in Guatemala the armed forces
split their support between two parties of similar ideological persuasions
but different personal loyalties (Batz, 2013). After the military coup of 1963,
b4407_Ch05.indd 105 1/6/2022 9:15:29 AM

6"×9"
Enrique Peralta Azurdia sought to consolidate the power of the military in

the electoral process by creating his own Institutional Democratic Party and
allowing the other party heavily infiltrated by the military to run in elections
(Handy, 1984, p. 158). A similar pattern was followed by the military of El
Salvador (Stanley, 2006).
The argument above implies that with respect to the mode of power
acquisition, monarchies and military regimes can be contrasted to all other
regime genomes as uniquely predisposed toward allowing greater degrees
of multiparty competition. Thus, there is a solid theoretical reason to expect
that high party system fragmentation, normally conjoined with high vol-
atility, will be positively associated with monarchies and military regimes.
With regards to monarchies, this association was quite noticeable already in
the patterns of party development of the 19th–early 20th century empires.
There is no ground to assume that the effects of monarchy and military
rule will be equal in strength, though. Royal control over the government, if
constitutionally enshrined, can be exercised irrespective of the composition
of legislative bodies. But from a political perspective, the presence of strong
parties, which leads to high party system concentration, effectively circum-
scribes the powers of the monarchy. Morocco provides an outstandingly clear
example of how a highly fragmented party system, with numerous parties
competing not only for the support of the electorate but also for the spoils
associated with loyalty to the royal ruler, can greatly enhance the monarch’s
effective powers (Cavatorta, 2009).
Military regimes are different in that while they are capable of neutral-
izing the political consequences of party system fragmentation, one can
hardly argue that excessive fragmentation is politically beneficial for them.
In the case of Guatemala, the ruling military’s failure to ensure the absence
of strong parties outside of their sphere of control ultimately contributed
to the demise of the regime (Johnson, 1971). In a similar vein, high party
system volatility can be tolerable or even beneficial for the executives in
monarchies, where constant rotation of political parties in the legislature
may weaken them and thus facilitate executive control over the party system.
b4407_Ch05.indd 106 1/6/2022 9:15:30 AM

But in military regimes, there is a stronger incentive to maintain stability

within the overtly pro-regime segment of the party system, be it represented
by a single pro-regime party or several such parties.
In democracies, party system volatility and nationalization are both
empirically associated with the effective number of parties: the former,
positively, because the electoral support of large parties naturally tends to be
more stable than that of small parties (Bartolini and Mair, 1990; Coppedge,
1998); and the latter, negatively, because highly fragmented party systems
often include parties relying primarily on the territorial bases of support,
while large parties tend to be nationalized (Golosov, 2016a). Such easily
predictable associations are not regime-specific. This implies that the party
systems of monarchies and military regimes will be more volatile and less
nationalized than those of other regime genomes.
The argument above can be summarized in the form of the following
hypotheses pertaining to the effects of two regime genomes, monarchy and
military regimes, upon party system parameters:
H1. The party systems of monarchies will be highly fragmented, highly

volatile, and lowly nationalized.
H2. The party systems of military regimes will be highly fragmented
and highly volatile in relation to all other political regime genomes,
but not in relation to the party systems of monarchies.
Since party system format is closely related to fragmentation, the above

hypotheses obviously imply that monarchies and military regimes will be
more supportive of multipartism than all other regime genomes. At the
same time, there is a theoretical reason to suggest that even in those cases
when the numbers of parties are small, monarchy will be better off with
fragmented opposition, while the military rulers often seek to enhance
their control over the electoral arenas by forcing all opposition politicians
to join a small number of parties licensed by the regime, as exemplified by
the paradigmatic cases of Brazil and Indonesia.
b4407_Ch05.indd 107 1/6/2022 9:15:30 AM

6"×9"
Party Legacies
Some of the authoritarian regimes inherit their political parties, or even
whole sets of parties, from the previous phases of their political develop-
ment. The resulting party systems can be viewed as exogenous in the sense
that the origins of political parties lie outside of the regime. Under different
authoritarian regimes, party systems form largely or even exclusively on the
basis of these regimes, in a relative isolation from the previous experiences.
Such party systems can be viewed as endogenous. The ways in which author-
itarian regimes can utilize party legacies are immanent for different varieties
of authoritarianism, which allows for viewing these ways as substantive,
rather than incidental, properties of political regime genomes. Therefore,
party legacies are an important element of the regime genome determinants
of authoritarian party systems. A similar pattern of inter-regime continuity
has been registered in research on the party systems of nascent democracies
(Remmer, 1985; Hicken and Kuhonta, 2011).
From this perspective, single-party and party electoral authoritarian
regime genomes stand out because in the former case, the formerly dominant
parties are almost invariably inherited, while in the latter case, party systems
can be inherited in their entirety, sometimes with only slight modifications
caused by the circumstances of regime transition. Dominant parties inherited
from the periods of single-party rule normally gain the lion’s share of seats
in the first authoritarian elections following the introduction of multiparty
competition, and it can be noticed — with the statistical proof provided in
the final section of this chapter — that the leading positions of these parties
tend to consolidate over time. The same applies to party electoral author-
itarian regimes. On the one hand, they do inherit opposition parties with
their niche electorates and experiences of parliamentary representation,
which certainly facilitates their survival and thus can be expected to reduce
the margins of victory of dominant pro-regime parties. But, on the other
hand, the levels of success gained by the permitted opposition parties in the
previous authoritarian contexts are not only small but also predetermined by
b4407_Ch05.indd 108 1/6/2022 9:15:30 AM

these contexts, so that they are likely to retain their stable but small niches
after the transition to a new form of electoral authoritarianism.
This is not necessarily the case with transitions from single-party rule
to multiparty elections because such transitions sometimes, albeit not
necessarily, are associated with greater levels of political uncertainty. For
example, the 1992 elections in Cameroon witnessed the entry of a strong
opposition party, the National Union for Democracy and Progress, led by
a former prime minister and utilizing his influence and popularity in some
of the regions of the country. The party was able to gain quite a significant
minority of seats in the first multiparty elections, leading to hopes for tran-
sition to democracy in Cameroon. By 1997, the regime of Paul Biya and
his Cameroon People’s Democratic Movement succeeded in consolidating
control over the election process, as a result of which the parliamentary
representation of the opposition was reduced to the margins (Gros, 1995),
but the uncertainty of the initial transition period certainly affected the
properties of the nascent authoritarian party system. Such uncertainties are
alien to transitions from one electoral authoritarian regime to another, as a
result of which the impact of the pre-existing parties can be inconsequen-
tial. However, there is no reason to expect that these two political regime
genomes will affect party system parameters in profoundly dissimilar ways.
The party legacies of democracies are different. Even in nascent democ-
racies that can be properly characterized as defective, which is normally the
case with those ultimately making transitions to authoritarian rule, political
parties rely upon societal bases of support and develop modes of inter-party
contestation and cooperation that can be sustainable over relatively long
periods of time (Kitschelt et al., 1999). At the same time, the party systems of
unconsolidated democracies tend to be characterized by high fragmentation
(Olson, 1998) and low nationalization stemming not only or even primar-
ily from the mechanical linkage between the nationalization of individual
parties and their size but rather from their failure to mobilize nationwide
electorates (Harbers, 2010). These properties, if inherited by authoritarian
party systems, can be uprooted. In fact, they do vanish over time due to the
b4407_Ch05.indd 109 1/6/2022 9:15:30 AM

6"×9"
changing nature of electoral politics. But some of the inherited parties are
resistant to change.
Consider the case of Russia. The Russian party system, as it emerged in
1993–1995, was highly fragmented and featured several parties with almost
exclusively regional appeal (Golosov, 1998, 2003b). But it did include two
relatively large parties, the Communist Party of the Russian Federation and
the Liberal Democratic Party of Russia, both standing in opposition to the
pro-reform policies of president Boris Yeltsin; several unstable formations
that either supported the government or were moderately critical of it; and a
plethora of fluid micro-parties. Most of them disappeared without leaving a
trace in the process of authoritarian transformation that ensued in the mid-
2000s, while pro-government groups were gathered under the umbrella of
the new dominant party, United Russia. However, the Communists and the
Liberal Democrats survived several attempts of hostile takeover by the exec-
utive (Gel’man, 2005) and managed to retain, albeit in a reduced form, their
linkages with specific societal (Schofield and Zakharov, 2010) and territorial
constituencies (Golosov, 2014d). Despite all efforts of the authorities, the
continuous presence of these survivors prevented United Russia from taking
outright majorities of the popular vote in the 2011 and 2016 parliamentary
elections. It is noticeable that for all the differences between Russia and
Guyana, the latter displayed a similar pattern of survival of parties inherited
from the democratic phase of political development (Premdas, 1978).
Military regimes stand in the middle ground because in most cases the
leaders of these regimes find it either undesirable or practically difficult to
rule without the assistance of civilian politicians who are normally affiliated
with pre-existing political parties (McKinlay and Cohan, 1975, p. 10). Even
in authoritarian Brazil, with its artificial two-party system being obviously
masterminded by the ruling military, the permitted political parties were
formed by absorbing the organizational residua of some of the parties that
existed before the military coup (Kinzo, 1988). There are exceptions. The
Golkar party of Indonesia was created and directly controlled by the mili-
tary without much participation of civilian politicians (Reeve, 1987), even
b4407_Ch05.indd 110 1/6/2022 9:15:30 AM

though the permitted opposition parties were formed in the same way as
in Brazil. The authoritarian party system of Mauritania comprised several
parties — including the pro-regime Democratic and Social Republican
Party — all of which, as noted by an observer, “had been founded out of
nothing” after the military came to power (Rebstock, 2005, p. 8). But most
of the military regimes listed in Table 5.1 did make extensive use of the
pre-existing political parties.
There are two regime genomes, monarchy and non-party electoral
authoritarian regime, that tend to generate endogenous party systems.
In the latter case, this can be assumed by definition. In the countries of
post-Soviet Central Asia, the emergence of pro-regime parties was preceded
by relatively long periods when their electoral arenas were dominated by
non-party-affiliated politicians most of whom were loyal to the dictatorial
executives (Golosov, 2020). The formation of dominant pro-regime parties
involved some contestation among different factions of these politicians,
as a result of which it was sometimes difficult for them to coalesce into a
single party even if all of them were loyal to the regime (Schatz, 2009). In
the course of time, however, nearly all these countries, with the exception
of Uzbekistan, did achieve political monopoly in the pro-regime segments
of their party spectra.
Monarchies, in theory, can inherit their authoritarian party systems,
but in practice, this happened in only one of the cases included in this
study, Nepal, where the liberalization of 1990 was preceded by a long
period of non-partisan elections under the so-called Panchayat system
(Khadka, 1986), but some of the important parties emerged well before the
introduction of multiparty elections. In Iran, the National Front coalition
of Mohammad Mosaddegh was outlawed after the Shah’s coup in 1953, so
that the authoritarian party system was created by recruiting previously
unaffiliated politicians or defectors from the National Front into entirely new
organizations (Westwood, 1961). In Morocco and Thailand, the formation
of political parties was a slow process that occurred entirely under the aus-
pices of monarchy and proceeded in two steps: first, gradual development
b4407_Ch05.indd 111 1/6/2022 9:15:30 AM

6"×9"
of opposition parties that normally lost elections to the pro-government

independents, and second, the formation of pro-regime parties, joined
mostly by these independents with some participation of defectors from
the opposition parties (Zartman, 1967; Neher, 1970).
Unlike in non-party electoral authoritarian regimes, the presence
of several pro-regime parties in monarchies tends to persist over time.
Nevertheless, the argument above clearly implies that the party systems of
non-party electoral authoritarian regimes will be more fragmented than
the party systems of all other regime genomes with the exception of mon-
archy and military regime. It has to be noticed that in the case of these two
regimes, the impact of party legacies is synergetic with the impact of the
mode of power acquisition.
The above analysis was focused mostly on the organizational legacies
of pre-existing political parties. However, political parties owe their election
results to voting behavior, and this is the second aspect of party legacies that
has to be taken into account. Gandhi and Lust-Okar (2009, p. 414) observe
that “what is striking about the literature to date is that electoral behavior in
authoritarian regimes is similar in many ways to that in democracies,” which
suggests that the modes of electoral mobilization are not regime-specific even
in juxtaposition between democracy and authoritarianism. Thus, the main
theories of voting behavior in democracies can serve as useful heuristics for
understanding similar behavioral patterns observed in authoritarian regimes.
In broad terms, the dominant approaches to voting behavior in democracies
can be divided into two categories, instrumental and expressive theories of
voting behavior (Tóka, 2009).
According to the expressive theories, voting is an act aimed primarily
at demonstrating political affinities, and the main benefit of voting for a
citizen stems from the intrinsic rewards of casting a vote (Tullock, 1971).
There is no reason to assume that expressive motivations are alien to the
authoritarian voter. The main instrument used to mobilize expressive voting
for autocracy is indoctrination. Traditionally, there used to be two often —
but not necessarily — interconnected tools of indoctrination, ideological
b4407_Ch05.indd 112 1/6/2022 9:15:30 AM

political parties (Gunther and Diamond, 2003) and mass propaganda in

the media. Ideological parties used to dominate the political landscapes
of single-party regimes, particularly those of the communist regimes,
and it is generally acknowledged that indoctrination constituted a core of
their interactions with the subject populations (Pop-Eleches and Tucker,
2014). The dominant parties of electoral authoritarian regimes lack ideo-
logical coherence and appeal (Isaacs and Whitmore, 2014). The decline of
party-based indoctrination, however, has been strongly compensated by a
drastic increase in the efficacy of media-based indoctrination, particularly
due to the development of the electronic media (Gehlbach et al., 2016),
and by the increased importance of the autocrat’s personal popularity
among the voters who may support the leader of the regime quite sincerely
(Frye et al., 2017).
The instrumental theories of voting posit that electors can punish or
reward governments on the basis of their performance, particularly in the
management of the national economy (Mueller and Stratmann, 2003).
Yet, the available evidence, however limited, suggests that the policy
effects of authoritarian elections are small (Malesky and Schuler, 2009).
At the same time, according to the “tragic brilliance” thesis advanced in
the literature (Diaz-Cayeros et al., 2003) and confirmed empirically in a
cross-national study (Golosov, 2016c), such regimes retain their ability
to achieve electoral success even in the conditions when the national
economy deteriorates.
This does not mean that instrumental incentives for voter turnout
are completely absent in authoritarian regimes. Yet, putting aside violent
coercion to vote, which is a highly resource-consuming practice confined
to a handful of particularly repressive regimes (Van Ham and Lindberg,
2016), this reduces such incentives’ viability to a very specific mode of
electoral behavior, clientelistic exchange. Clientelism is a way of electoral
mobilization based on trading particularistic (material) benefits to voters
in exchange for their votes (Stokes, 2000; Kitschelt and Wilkinson, 2007).
Political organizations that mobilize their voters primarily by means of
b4407_Ch05.indd 113 1/6/2022 9:15:30 AM

6"×9"
clientelistic exchange are referred to as political machines. It is possible

to say that political machines represent the apex in the development of
clientelistic exchange. Clientelism is widespread, both across the world
and over time, but mature political machines are characteristic of rela-
tively few countries, and by far not all of them are autocracies. In fact,
the archetypal historical cases of machine politics had been observed in
some of the large cities of the United States in the 19th–early 20th century
(Shefter, 1994).
The available body of in-depth research on electoral politics in
authoritarian regimes, especially in the Middle East and Latin America,
registers endemic clientelism with a striking regularity (Magaloni, 2006;
Blaydes, 2010). As noticed by Lust-Okar (2006, p. 456), “elections under
authoritarianism are an important arena for competition, but one over
patronage, not policy.” But, given that clientelism is not even specific to
authoritarian rule, is there any reason to suggest that there may be a sig-
nificant variation on this parameter among authoritarian political regime
genomes? Not necessarily. In those conditions when clientelistic exchange
reaches the level of fully developed machine politics, there are all reasons
to expect that the effects of expressive voting and clientelistic exchange
will converge in their impact upon party system properties. The founda-
tions of patron–client relationships lie at the level of very personal — one
would even say, intimate — ties between the two parties of the deal (Scott,
1972). But in machine politics, direct face-to-face interactions between
patrons and clients are likely to be replaced with chains of broker rela-
tionships (Stokes, 2007). This is why, despite the deeply localized nature
of clientelistic politics in authoritarian Mexico, which has been registered
in many studies (Magaloni et al., 2007; Hilgers, 2008), the party system
of the country featured continuously high levels of nationalization and
moderate levels of volatility.
Armenia and Russia, in their capacities of post-democracies, represent a
different pattern. The wide spread of machine politics in some of the regions
of Russia contributed to the entrenchment of the political monopoly of the
b4407_Ch05.indd 114 1/6/2022 9:15:30 AM

United Russia party and, ultimately, in similarity to what has been w itnessed
in Mexico, to a very high level of nationalization of the party system (Golosov,
2015c; Tkacheva and Golosov, 2019). But the level of volatility in the devel-
opment of Russia’s authoritarian system, while certainly lower than during
the period of its experimentation with democracy, was notably high. The
same is true with regards to the highly clientelistic (Baghdasaryan, 2017)
but rather volatile pattern of party politics in Armenia.
The explanation for this phenomenon can be inferred from the con-
ceptual model of the transferability of political machines as developed by
Golosov (2014d). In post-democracies, authoritarian party systems form
by absorbing the organizational residua of parties that have operated in
democratic conditions, co-opting most viable of these residua under the
umbrella of the dominant pro-regime parties (Golosov, 2014b). For such
parties, political machines are a very valuable asset, which explains why
the regional elites who control the machines are the primary targets of
cooptation irrespective of their previous party affiliations (Reuter, 2010). But
when political machines, with their massive capacities for voter mobilization,
are transferred from one party to another, this necessarily manifests itself in
increased levels of volatility. The case of Guyana is deviant because the ethnic
bases of the country’s political parties (Despres, 1975) made them highly
resistant to the transferability of political machines, so that the volatility of
the authoritarian party system remained low.
The argument above can be summarized in the form of the following
hypotheses pertaining to the effects of the regime genomes upon party
system parameters:
H3. The party systems of party electoral authoritarian regimes will be

lowly fragmented, lowly volatile, and highly nationalized.
H4. The party systems of single-party regimes will be lowly fragmented
and lowly volatile in relation to all other political regime genomes,
but not in relation to the party systems of party electoral authori-
tarian regimes.
b4407_Ch05.indd 115 1/6/2022 9:15:30 AM

6"×9"
H5. The party systems of non-party electoral authoritarian regimes will

be more fragmented than the party systems of military, single-party,
and party electoral authoritarian regimes.
H6. The party systems of post-democracies will be highly volatile.
With regards to party system format, the party legacies of all political
regime genomes, with the exceptions of monarchy and military regimes,
are hardly conducive to multipartism, even though the residual elements of
multiparty competition are likely to be observed in post-democracies during
the formative phases of their authoritarian party systems. In party systems
inherited from single-party regimes and non-party electoral authoritarian
regimes, we can expect a propensity toward greater fragmentation among the
permitted opposition. In both cases, this can be related to the high levels of
uncertainty stemming from the profound change experienced by these party
systems during their formative phases. In party systems inherited from non-
party electoral authoritarian regimes, fragmentation within their opposition
segments will be lower, thus moving these systems closer to the bipartite model.
Empirical Methods
In order to test the above-formulated hypotheses empirically, I use several
methods depending on the scope and characteristics of the data at hand.
Most extensive tests are performed on the continuous dependent variables
with sufficiently large numbers of election-level observations, party system
fragmentation, operationalized as the Golosov effective number of parties,
and volatility, operationalized as the Pedersen index calculated on the basis
of seats. Justifications for these operational definitions are provided in
Chapter 4. Each of the political regime genomes is coded on a binary basis,
which leaves me with six dichotomous variables. The units of analysis are
defined at the election level.
I start with the descriptive statistics, thus obtaining a preliminary
understanding of how the levels of fragmentation and volatility are related
b4407_Ch05.indd 116 1/6/2022 9:15:30 AM

to different regime genomes. Since there are theoretical reasons to expect

that some of the genomes exert similar/converging influences upon the
party system parameters of interest, I use the descriptive statistics in order to
identify such similar influences and create additional independent variables
based on groups of the original categories. Later in this analysis, the prod-
ucts of such grouping are referred to as the combined types. The theoretical
discussion above has demonstrated that the influences of different regime
genomes upon fragmentation and volatility do not necessarily concur. For
this reason, the combined types are defined separately for the statistical tests
performed on each of the two dependent variables.
As the first step in the statistical analysis proper, I run ANOVA multiple
comparisons of means with Bonferroni post hoc tests. The tests are performed
on the sets of dependent variables that include categories pertaining to the
combined types and the remaining individual types. As the second step, in
order to take into account the country-level clustering in the data and, more
importantly, with the aim of introducing necessary control variables, I run
standard linear regressions with clustered-robust standard errors (n = 294
and 237 for fragmentation and volatility, respectively; number of clusters =
57). Multiple regression analyses are performed separately for each of the
regime genomes, including the combined ones, by treating them as dichot-
omous variables.
As the third step, I run multilevel mixed-effects regressions. This allows
for a different way of accounting for country clustering in the data and adds
the value of accounting for time affects as well. More importantly, in these
analyses, unlike in standard linear regressions, the effects of different regime
genomes are juxtaposed with each other. The effects of monarchy are not
estimated at this third stage not only because they are clear enough from
the previous analyses but also in order to avoid collinearity. Note that while
in the tables of descriptive statistics the values of volatility are reported as
percentages, all regression analyses were performed on the decimal shares.
All multivariate analyses include identical sets of control variables. The
substantive justification for the inclusion of these particular control variables
b4407_Ch05.indd 117 1/6/2022 9:15:30 AM

6"×9"
is thus: First, there are reasons to expect that the harshest forms of political
monopoly are likely to emerge in highly autocratic political settings, while
relatively liberalized authoritarian regimes will be associated with party
systems that are more fragmented, because greater number of parties are
allowed to run in elections, and more volatile because of the relative ease
of electoral entry. Therefore, I expect that the level of democracy will be
positively associated with both of the dependent variables. Consistent with
the overall approach of this book, the level of democracy is operationalized
as the V-Dem index of electoral democracy for the years of holding the
included elections. Note that in democracies, the direction of the empirically
observed association between the level of democracy and party system frag-
mentation is positive, but only if the durability of democracy is controlled
for (Golosov, 2015b). In fact, it is traditional in political science to argue
that mature democratic party systems are less fragmented and less volatile
than those of inchoate democracies (Mainwaring, 1999).
Second, the foundational literature on the structural determinants of
democratic party systems suggests that in societies that are cleaved along eth-
nic or linguistic lines, party systems tend to be fragmented, particularly due
to the salience of regional parties (Ordeshook and Shvetsova, 1994; Amorim
Neto and Cox, 1997). The available empirical evidence is inconclusive in the
sense that ethnic diversity has been found to be positively associated with
party systems’ fragmentation, but the effect is strongly contingent on the
electoral rules (Clark and Golder, 2006; Singer and Stephenson, 2009; Moser
and Scheiner, 2012; Lublin, 2017b) or other institutional factors and societal
characteristics (Stoll, 2013). At least in some contexts, the association tends
to be nonlinear (Raymond, 2015). It is also important to mention that ethnic/
linguistic diversity is sometimes considered as proxy for the general num-
ber of social cleavages, even though there are alternatives to this obviously
restrictive approach (Riera, 2020). Yet, given the paramount importance
attributed to this factor, no research on party system parameters can afford
omitting it from the list of control variables. For operationalization, I use
the measure of ethnic diversity developed by Fearon (2003).
b4407_Ch05.indd 118 1/6/2022 9:15:30 AM

Third, I control for the level of economic development, operationally

defined as real GDP per capita in 2011 USD equivalent at the time of elec-
tions. The data are derived from the dataset developed by the Maddison
Project (Bolt and van Zanden, 2020). Similarly to the previously described
control variable, GDP per capita is often used as proxy for the overall level
of societal maturity, which implies, albeit not without serious caveats, that
in democracies the probability of stable patterns of party competition
increases in association with higher levels of economic development (Lane
and Ersson, 2007). A recent empirical study concludes that socio-economic
development is associated with higher levels of party system fragmentation
(Raymond, 2018). At the same time, it can be argued that under authoritar-
ianism, higher levels of economic development lead to the enhanced state
capacity and thereby, can be associated with less fragmented party systems
(Seeberg, 2014). Even without any specific expectations regarding the possi-
ble direction of causality, or lack of it thereof, it is therefore useful to include
GDP per capita as a control variable. Similar considerations, supported by
previous research on democracies (Anckar, 2000) and theoretical consid-
erations (Sikk and Taagepera, 2014) led me to the inclusion of population
size at the time of elections, also as reported by the Maddison Project, as
the fourth control variable. However, the impact of this factor in all models
turned out so negligibly small that the related coefficients are not reported.
Finally, I use a control variable that pertains to the specifics of the data
at hand. As any social phenomena, authoritarian party systems evolve over
time. The overall impact of temporal dynamics is estimated in the multilevel
models, but they do not account for the effects of certain moments in party
system development, sometimes referred to as critical junctures (Ufen, 2012),
upon the dependent variables. Specifically, I expect that in the development
of authoritarian party systems, the second elections constitute the point at
which, to a greater extent than earlier or later, they liberate themselves from
those legacies of the previous state of party politics that are inconsistent with
the properties of the emerging authoritarian regime. This implies a decline
in fragmentation and a rise in volatility. In order to estimate these effects,
b4407_Ch05.indd 119 1/6/2022 9:15:30 AM

6"×9"
Table 5.5. The descriptive characteristics of the control variables.

Standard
Index of electoral democracy 0.25 0.24 0.06 0.40 0.07
Index of ethnic diversity 0.48 0.48 0.00 0.93 0.25
Real GDP per capita in 2011 6.66 4.00 0.69 65.66 8.44
USD equivalent (thousands)
Second elections 0.19 0.00 0.00 1.00 0.40
I employ a dichotomous variable that codes second elections in each of the

country-level sequences as 1, and all other elections, as 0. The descriptive
characteristics of the control variables are reported in Table 5.5.
Unfortunately, this complex set of statistical tools cannot be applied to
party system volatility and nationalization. As demonstrated in Chapter 4,
party system format is best revealed by means of a graphic representation,
the relative-size triangle, and such graphic representations will be provided
for each of the regime genomes. The graph can also be used for allocating
each of the party systems to one of the six categories, ranging from systems
with outright one-party domination to balanced multiparty systems, in
accordance with the quantitative criteria set in Table 4.5. This opens the
possibility of a more rigorous quantitative investigation in the form of the
two-way analysis of variance.
The analysis of the determinants of party system nationalization
is severely hindered by the limited availability of the data. With only
42 country-level observations at hand and six possible sources of causality,
complex statistical methods can yield only unstable results. This made me
limit my inquiry to descriptive statistics and establishing pairwise correla-
tions between the dichotomous variables pertaining to political regime
genomes and the index of party system nationalization.
Empirical Findings
The empirical findings presented in Tables 5.6–5.10 pertain to the effects of
political regime genomes upon party system fragmentation. The descriptive
b4407_Ch05.indd 120 1/6/2022 9:15:30 AM

Table 5.6. Party system fragmentation by regime genomes.

Standard
Monarchy 4.76 4.29 1.16 11.80 3.20 21
Military regime 1.83 1.57 1.00 4.59 0.79 63
Single-party regime 1.35 1.23 1.00 2.97 0.39 98
Post-democracy 2.00 1.98 1.29 4.35 0.87 11
Party electoral authoritarian 1.36 1.24 1.00 2.31 0.36 80
regime
Non-party electoral 1.93 1.63 1.00 3.51 0.81 21
Post-democracy + non-party 1.96 1.66 1.00 4.35 0.82 32
electoral authoritarian regime
Single-party regime + party 1.36 1.23 1.00 2.97 0.38 178

by fragmentation.
Between groups 220.14 3 73.38 73.46 0.00
Within groups 289.69 290 1.00 — —
Total 509.83 293 — — —
data provided in Table 5.6 is illustrative enough, but its primary purpose is

to identify the bases for creating two combined categories, defined as “Post-
democracy + non-party electoral authoritarian regime” and “Single-party
regime + party electoral authoritarian regime.” As is evident from Tables 5.7
and 5.8, the differences among political regime genomes on the parameter
of party system fragmentation are quite significant. Monarchy stands out
as a political regime that is associated with most fragmented party systems,
while the combined category of single-party regimes and party electoral
authoritarian regimes generates party systems that are significantly less
fragmented than can be registered for all other genomes. This fully con-
firms the above-stated hypotheses, from H1 to H4, with regards to party
b4407_Ch05.indd 121 1/6/2022 9:15:30 AM

6"×9"

genomes by fragmentation.
Mean Standard
Monarchy Military 2.94 0.25 0.00
Single-party regime + 3.41 0.23 0.00
party electoral
Post-democracy + 2.81 0.28 0.00
non-party electoral
Military Monarchy −2.94 0.25 0.00
party electoral
Post-democracy + −0.13 0.22 1.00
non-party electoral
Single-party Monarchy −3.41 0.23 0.00
regime + Military −0.47 0.15 0.01
Party electoral
authoritarian Post-democracy + −0.60 0.19 0.01
regime non-party electoral
Post-democracy + Monarchy −2.81 0.28 0.00
Non-party electoral Military 0.13 0.22 1.00
authoritarian
regime Single-party regime + 0.60 0.19 0.01
party electoral
system fragmentation. Note that in accordance with my theoretical expec-

tations, the party systems of monarchies are significantly more fragmented
than those of military regimes. The lack of significant difference
between military regimes and non-party electoral regimes is consistent
with H5.
The findings reported in Tables 5.7 and 5.8 allow for the fol-
lowing schematic presentation of political regime genomes in the
b4407_Ch05.indd 122 1/6/2022 9:15:30 AM

Table 5.9. Linear regression results, effective number of parties by

political regime genomes (n = 294 adjusted for 57 clusters; unstandardized
coefficients; clustered-robust errors in parentheses).
Single-party
regime + party Post-democracy +
electoral non-party electoral
authoritarian authoritarian
Monarchy Military regime regime
Political regime 3.28** (1.32) 0.06 (0.32) −1.08*** (0.41) 0.21 (0.32)
genomes
Index of electoral 2.61 (1.99) 2.21* (1.22) 2.67** (1.07) 2.10* (1.26)
democracy
Index of ethnic −0.57* (0.30) −0.13 (0.28) 0.08 (0.28) −0.10 (0.28)
diversity
Second elections −0.17 (0.07) −0.19** (0.09) −0.29*** (0.11) −0.20** (0.09)
Real GDP per −0.01 (0.01) −0.02* (0.01) −0.01* (0.01) −0.02*** (0.01)
capita in 2011
USD equivalent
(thousands)
Constant 1.23*** (0.43) 1.34*** (0.40) 1.86*** (0.41) 1.35*** (0.34)
R-squared 0.43 0.03 0.18 0.03
Note: *** significant at 0.01 level, ** significant at 0.05 level, and * significant
at 0.1 level.
descending order of their positive impact upon party system volatility:

Monarchy > (Post-democracy + Non-party electoral authoritarian
regime) ≥ Military regime > (Single-party regime + Party electoral author-
itarian regime), where > and ≥ stand for significant and insignificant
differences, respectively. This ordering clarifies the relative strength of effects
predicted in H2 and H5 by indicating that in this respect, party legacies are
more important than modes of power acquisition. Table 5.9 suggests that
while the effects of monarchy and the combined category of single-party
regimes and party electoral authoritarian regimes are significantly associated
with high party system fragmentation, those of two other categories are not.
However, the results of the multilevel regression, as reported in Table 5.10,
show that if the effects of monarchy are not taken into account, all three
b4407_Ch05.indd 123 1/6/2022 9:15:30 AM

6"×9"

number of parties by political regime genomes.
Coefficient Standard error Significance
Military −2.45 0.48 0.000
Single-party regime + party electoral −2.91 0.45 0.000
Post-democracy + non-party −2.41 0.53 0.000
Index of electoral democracy 1.97 0.94 0.036
Index of ethnic diversity −0.45 0.45 0.318
Second elections −0.19 0.10 0.053
Real GDP per capita in 2011 USD −0.01 0.01 0.406
equivalent (thousands)
Constant 4.11 0.54 0.000
Random effect parameters (estimate/standard error)
Country 0.79 0.09 —
Year 0.00 0.01 —
Note: Wald chi2 = 50.02; Log restricted-likelihood = −346.27; Prob >

chi2 = 0.000.
remaining categories turn out to be negatively associated with the dependent

variable. Note that the values of the coefficients reported in Table 5.10 are
fully consistent with my theoretical expectations.
Among the control variables, the index of ethnic diversity can be
assessed as almost entirely unrelated to the fragmentation of authoritarian
party systems. It is however noticeable that in contrast to the theoretical
expectations derived from the experience of democracies, the observed
associations, for all their insignificance, are invariably negative. Ethnically
diverse autocracies tend to have relatively concentrated party systems. The
significance of the level of economic development, as expressed in GDP per
capita, is variable but noticeable in most of the models reported in Table 5.9.
The registered impact is always negative, which is consistent with the above-
cited argument regarding the relationship between state capacity and party
b4407_Ch05.indd 124 1/6/2022 9:15:30 AM

system fragmentation under authoritarian rule. In full accordance with my

expectations, the index of electoral democracy tends to be positively asso-
ciated with the dependent variable, even though the associations are rather
weak. My reasoning regarding the impact of the second elections also finds
a confirmation in the results of the statistical tests, but the observed effects
are also weak even if statistically significant.
Figure 5.1 and Table 5.11 clarify the findings reported above with
regards to party system format. As is evident from the figure, constellations
without a majority party are relatively rare, which explains why Table 5.11
merges them all into a single category. The analysis reported in the table
is consistent with the previous findings by demonstrating that monarchies
are strongly positively associated with multipartism, and strongly negatively
with both formats where there are majority parties. The positive association
between multipartism and military regimes is weaker but also evident. At
the same time, the analysis shows that among the predominant party sys-
tems, high fragmentation within the opposition camp is not characteristic
of military regimes, which is consistent with my expectations. Single-party
regimes and party electoral authoritarian regimes are similar in that they
are negatively associated with multipartism, but if the former category is
strongly predisposed toward predominant party system with fragmented
opposition, the latter displays a weak but positive association with the
systems with concentrated oppositions. This, in the theoretically expected
way, contrasts them to non-party electoral authoritarian regimes where the
directions of associations are opposite: strongly positive with the systems
with high opposition fragmentation, and strongly negative with the systems
with concentrated oppositions.
The empirical findings presented in Tables 5.12–5.16 pertain to the
effects of political regime genomes upon party system volatility. The descrip-
tive statistics in Table 5.12 allows for identifying two combined categories,
“Military Regime + Post-Democracy” and “Single-Party Regime + Party
Electoral Authoritarian Regime.” The latter category, in accordance with
b4407_Ch05.indd 125 1/6/2022 9:15:30 AM

6"×9"
1 1
0.9
Plate 1 0.9
Plate 2
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 1
Plate 3 0.9
Plate 4
0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 1
Plate 5 0.9
Plate 6
0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

authoritarian elections by political regime genomes. Plate 1: Monarchy.
Plate 2: Military regime. Plate 3: Single-party regime. Plate 4: Post-
democracy. Plate 5: Party electoral authoritarian regime. Plate 6: Non-
party electoral authoritarian regime.
H3 and H4, is the same as one of those employed in the above analysis
of party system fragmentation, while the former is defined in a different
way, which accords with H6. Tables 5.13 and 5.14 show that the differences
among the political regime genomes on the parameter of volatility are far
b4407_Ch05.indd 126 1/6/2022 9:15:30 AM

Table 5.11. Two-way analysis of variance: regime genomes and party

system formats.
Predominant
party system Other
with fragmented systems, Pearson Cramér’s Fisher’s
opposition, yes/no yes/no chi2 V exact
Monarchy 1/110 20/163 10.48*** −0.19 0.00
Military regime 17/94 46/137 3.96** −0.12 0.06
Single-party regime 47/64 51/132 6.51** 0.15 0.02
Post-democracy 3/108 8/175 0.53 −0.04 0.54
Party electoral 30/81 50/133 0.00 0.00 1.00
Non-party electoral 13/98 8/175 5.61** 0.14 0.03
Predominant
party system with Other
concentrated systems,
opposition, yes/no yes/no
Monarchy 3/109 18/164 5.44** −0.14 0.02
Military regime 23/89 40/142 0.09 −0.02 0.88
Single-party regime 43/69 55/127 2.08 0.08 0.16
Post-democracy 5/107 6/176 0.26 0.03 0.75
Party electoral 37/75 43/139 3.10* 0.10 0.08
Non-party electoral 1/111 20/162 10.66*** −0.19 0.00
Systems without Other
majority party, systems,
yes/no yes/no
Monarchy 17/54 4/219 39.84*** 0.37 0.00
Military regime 23/48 40/183 6.69*** 0.15 0.01
Single-party regime 8/63 90/133 20.51*** −0.26 0.00
Democracy 3/68 8/215 0.06 0.01 0.73
Party electoral 13/58 67/156 3.74* −0.11 0.07
Non-party electoral 7/64 14/209 1.04 0.06 0.30
at 0.1 level.
b4407_Ch05.indd 127 1/6/2022 9:15:30 AM

6"×9"
Table 5.12. Party system volatility by regime genomes.

Standard
Monarchy 24.82 21.33 2.88 58.21 12.74 17
Military regime 19.76 16.17 0.00 64.99 15.06 49
Post-democracy 17.40 15.17 3.77 42.29 11.75 8
Party electoral authoritarian 12.10 7.41 0.00 52.27 13.52 69
regime
Military regime + 14.96 12.06 0.00 64.99 15.18 74
post-democracy
Single-party regime + party 10.38 5.23 0.00 56.79 12.86 178
electoral authoritarian
regime

by volatility.
Degrees
Sum of of Mean
Between groups 0.36 3 0.12 6.82 0.00
Within groups 4.13 233 0.02 — —
Total 4.49 236 — — —
less pronounced than those observed in the case of fragmentation, but some
of them are significant. Consistent with H1 and H2, the party systems of
monarchies, on the one hand, and military regimes and post-democracies,
on the other hand, are volatile. In this respect they are significantly different
from the party systems of single-party regimes and party electoral author-
itarian regimes, but not from those of non-party electoral authoritarian
regimes. This generally confirms H3, H4, and H5 with regards to party
system volatility.
b4407_Ch05.indd 128 1/6/2022 9:15:31 AM


genomes by volatility.
Mean Standard
Monarchy Military + post-democracy 0.05 0.04 0.86
party electoral
Military + Monarchy −0.05 0.04 0.86
post-democracy Single-party regime + 0.07 0.02 0.01
party electoral
Single-party Monarchy −0.12 0.03 0.00
regime +
Military + post-democracy −0.07 0.02 0.01
party electoral
authoritarian Non-party electoral −0.02 0.04 1.00
regime authoritarian regime
Non-party Monarchy −0.11 0.05 0.14
electoral
authoritarian
regime Single-party regime 0.02 0.04 1.00
+ party electoral
The findings reported in Tables 5.13 and 5.14 allow for the following
schematic presentation of political regime genomes in the descending order
of their positive impact upon party system volatility: Monarchy ≥ (Post-
democracy + Military regime) ≥ Non-party electoral authoritarian regime
≥ (Single-party regime + Party electoral authoritarian regime), where ≥
stands for the directions of differences, respectively. All differences among
the categories that are adjacent on the scheme are statistically insignificant,
but, as is evident from the tables, some of the differences between non-
adjacent categories are significant. Table 5.15 suggests that while the effects
b4407_Ch05.indd 129 1/6/2022 9:15:31 AM

6"×9"
Table 5.15. Linear regression results, volatility by political regime

genomes (n = 237 adjusted for 57 clusters; unstandardized coefficients;
clustered-robust errors in parentheses).
Single-party
regime + party Non-party
electoral electoral
Military + post- authoritarian authoritarian
Monarchy democracy regime regime
Political regime 0.10** (0.05) 0.06* (0.03) −0.07*** 0.003 (0.028)
genomes (0.02)
Index of electoral 0.40** (0.18) 0.38** (0.16) 0.42*** 0.39** (0.18)
democracy (0.16)
Index of ethnic 0.01 (0.04) 0.04 (0.04) 0.03 (0.04) 0.02 (0.04)
diversity
Real GDP per −0.002* (0.001) −0.002* (0.001) −0.002 −0.003* (0.001)
capita in 2011 (0.001)
USD equivalent
(thousands)
Second elections 0.06*** (0.02) 0.05*** (0.02) 0.05*** 0.06*** (0.02)
(0.02)
Constant 0.05 (0.05) 0.04 (0.04) 0.09** (0.04) 0.05 (0.05)
R-squared 0.14 0.14 0.17 0.11
Note: *** significant at 0.01 level, ** significant at 0.05 level, and * significant at 0.1 level.
of monarchy and the combined category of single-party regimes and party

electoral authoritarian regimes are significantly associated with party system
volatility, those of two other categories are not. However, the results of the
multilevel regression, as reported in Table 5.16, show that the effects of all
regime genomes, with the exception of the combined category of single-party
regimes and party electoral authoritarian regimes, fade away if monarchy is
removed from the model. The values of the coefficients reported in Table 5.16
are consistent with my theoretical expectations, though.
Among the control variables, the impact of the index of ethnic diversity
remains insignificant but positive in all models, which replicates the pattern
observed in the analysis of fragmentation. In a similar vein, the significance
of the level of economic development, as expressed in GDP per capita,
is variable but noticeable in most of the models reported in Table 5.16.
b4407_Ch05.indd 130 1/6/2022 9:15:31 AM

Table 5.16. Multilevel mixed-effects regression results, volatility by

political regime genomes.
Standard
Coefficient error Significance
Single-party regime + party −0.09 0.05 0.070
Non-party electoral authoritarian −0.07 0.06 0.269
regime
Index of ethnic diversity 0.02 0.05 0.718
Second elections 0.04 0.02 0.023
Constant 0.13 0.06 0.040
Country 0.07 0.01 —
Year 0.07 0.02 —
Note: Wald chi2 = 17.60; Log restricted-likelihood = 145.01; Prob > chi2 = 0.01.
The registered impact is always negative. The index of electoral democracy

emerges as the second strongest positive predictor of authoritarian party
system volatility, which fully confirms my theoretical expectations. But the
most systematic, highly statistically significant positive impact upon the
dependent variable is exerted by the second elections variable. This shows
that to a greater degree than fragmentation, party system volatility is con-
tingent upon the situational political contexts of elections rather than upon
structural factors.
The empirical findings presented in Tables 5.17 and 5.18 pertain to the
effects of political regime genomes upon party system nationalization. The
results of the analysis confirm H1 and H3, as the levels of nationalization of
the party systems of monarchies are indeed low, while party electoral author-
itarian regimes do have highly nationalized party systems, even though in the
latter case, the observed association is on the brink of statistical significance.
b4407_Ch05.indd 131 1/6/2022 9:15:31 AM

6"×9"
Table 5.17. Party system nationalization by regime genomes.

Standard
Monarchy 0.49 0.46 0.29 0.72 0.22 3
Military regime 0.74 0.77 0.50 0.90 0.15 10
Post-democracy 0.61 0.61 0.56 0.66 0.07 2
Party electoral 0.81 0.78 0.69 0.94 0.09 8

nationalization by political regime genomes, n = 42.
r Significance
Monarchy −0.35 0.02
Military 0.08 0.60
Single-party regime −0.06 0.69
Democracy −0.13 0.41
Party electoral authoritarian regime 0.26 0.10
Non-party electoral authoritarian regime 0.05 0.76
It is interesting to note that, albeit in a statistically insignificant way, the party

systems of post-democracies are negatively associated with party system
nationalization. This is consistent with the well-grounded observation that
some of the parties inherited from the democratic phases of these countries’
political development survive due to the presence of the regional bases of
electoral support (Gilley, 2010; Golosov et al., 2016; Reisinger and Moraski,
2017), even though, due to the small number of observations, this finding
may be unstable.
b4407_Ch05.indd 132 1/6/2022 9:15:31 AM

Chapter 6
Authoritarian Party Systems,

Executive–Legislative
Relations, and Federalism
T
he notion of political regime genome, as introduced in Chapter 5,
clearly implies that the autocratic rulers are not in a position to
define some of the fundamental characteristics of the regime over
which they preside. A military junta comes to power without having an
option of acting as a royal family or a mass political party, and even if, in
the course of time, a leader of the junta chooses to ascend to the throne
or create a party, the military origins of the regime remain consequential.
Thus, regime genomes are exogenous to regimes themselves. This is not the
case with the formal institutional structures of authoritarian regimes. Many
authoritarian regimes create entirely new institutional settings. Even if they
do not, they can easily modify the inherited institutional structures in a way
that facilitates their survival in power. As a result, the formal institutional
structures of authoritarian regimes are largely endogenous.
Of course, authoritarian political institutions are not unique in this
respect. The problem of endogenous institutions is inherent in contempo-
rary social research. As elegantly summarized by Przeworski (2004, p. 527),
“institutions are endogenous: their form and their functioning depend on
the conditions under which they emerge and endure. Now, the embarrass-
ingly obvious observation is that if endogeneity is strong, then institutions
cannot have a causal efficacy of their own.” For a very apparent reason, this
kind of endogeneity is stronger in autocracies than it is in democracies,
133
b4407_Ch06.indd 133 1/6/2022 9:15:45 AM

6"×9"
but it has to be recognized that institutional endogeneity is present in all

political regime settings. This does not undermine the research agenda of
neo-institutionalism in political science in general (Peters, 2019); nor does it
invalidate the recent institutional turn in the comparative studies of authori-
tarianism (Pepinsky, 2014). But it does require paying careful attention to the
incentive structures lying beyond endogenous institutions and their political
consequences (Rodden, 2009). Without stating any specific expectations at
this stage, it is important to emphasize that there is no reason to anticipate
that under authoritarianism, these incentive structures will be the same as
in democracies, even if institutions themselves look strikingly similar.
Indeed, it is unusual for contemporary autocracies to develop institu-
tional structures that are significantly different from those of democracies.
The extinct Libyan Jamahiriya and the current Islamic Republic of Iran are
rare exceptions. This chapter deals with the same categories of institutional
arrangements that are typically dealt with in political science in general, and
particularly in the voluminous literature on the institutional determinants of
party systems in democracies: patterns of executive–legislative relations and
federalism. The first section briefly discusses some of the problems in the
taxonomy of executive–legislative relations in application to authoritarian-
ism. The second section assigns authoritarian party systems to the different
types of institutional arrangements, lays out my theoretical expectations
and hypotheses, and then tests them statistically. The third section deals in
the same way with federalism. It has to be mentioned that in this book I do
not address one institutional dimension that is presumably important in
democracies, bicameralism. A recent study convincingly demonstrates that
the effects of this provision in autocracies are negligible, and the incentives
for its adoption can be reduced to the cooptation of sub-national elites and
to what the authors refer to as “masking controversial reforms” (Baturo and
Elgi, 2018, p. 919).
In order to provide for consistency in my analysis, I use the same set of
analytical tools as in Chapter 5, which makes it unnecessary to carve out a
separate section on methodology. In particular, it is important to mention
b4407_Ch06.indd 134 1/6/2022 9:15:45 AM

Political Institutions and Authoritarian Party Systems 135
that the set of control variables employed for testing some of the hypothe-
ses is exactly the same as previously. Of course, the observed effects of the
control variables are very similar to those reported in Chapter 5, which
liberates me from the necessity to discuss them separately, even though
the analysis takes these effects into account. The hypotheses are numbered
throughout the chapter.
Types of Executive–Legislative Relations

Among the patterns of executive–legislative relations that can be found in
contemporary autocracies, monarchy stands out as the only form of institu-
tional design that fully overlaps with a political regime type. Formally, some
of the monarchies can be placed into the category of parliamentary systems,
but it is clear that if the royal executive powers are significant, as observed
in the cases of Iran, Morocco, Nepal, and Thailand included in this study,
then the norm of government responsibility to parliament cannot be met
in full. In fact, the origin of the contemporary parliamentary system lies in
“the shift of the power to appoint governments from the crown to elected
assemblies” (Przeworski et al., 2012, p. 101). The effects of monarchy upon
authoritarian party systems have been discussed in Chapter 5 of this book,
which explains why these effects are addressed only in passing, if at all, in
this chapter, even though the statistical findings pertaining to these effects
are reported.
All other major forms of executive–legislative relations can be found
both in democracies and in autocracies, as long as they formally recognize
the principle of popular sovereignty and conduct elections. Of course, the
main distinction has been traditionally drawn between parliamentary and
presidential forms of government, the two prevailing models that were
devised and put into practice as early as in the 18th century (Blondel and
Thiebault, 2020, p. 169). It would be fair to say that this distinction remains
the focus of the ongoing cross-national research on executive–legislative
relations (Elgie, 2005; Samuels and Shugart, 2010).
b4407_Ch06.indd 135 1/6/2022 9:15:45 AM

6"×9"
The presidential form of government, or presidentialism, is defined

primarily by the presence of the chief executive who is elected directly, i.e.,
by popular vote, and does not depend on legislative confidence but rather
sits for a fixed term (Lijphart, 1984, pp. 66–71). The notion of chief execu-
tive implies that she names and directs the composition of the government,
which is explicitly recognized in the set of criteria developed by Shugart and
Carey (1992, p. 19). They add that in order to possess effective executive
powers, the president has to have some constitutionally granted law-making
authority. Thus, there are two directly elected institutions, the president and
the assembly, both of which possessing significant powers. In parliamen-
tary systems, often referred to as parliamentarism, there is only one such
institution, the assembly. The chief executive who names and directs the
composition of the government is the prime minister who is determined
by the assembly. The head of state, be it a monarch or a president, performs
mostly ceremonial functions. The terms of the assembly, and thereby of the
chief executive and her government, are not fixed.
In the relatively recent literature, the dichotomy between presiden-
tialism and parliamentarism has been complemented by two other types
of executive–legislative relations. One of them, by far more salient than
the other, is semi-presidentialism. The category was introduced in a series
of French-language books in the 1970s and then widely popularized by
Duverger (1980). The operational criteria used in the original formulation
were not entirely clear, which led Shugart and Carey (1992) to an outright
rejection of the concept in favor of a more delicate distinction between
president-parliamentary and premier-presidential systems. Later, however,
Shugart (2005) conceded that the concept did have merits of its own. A major
contribution to the study of semi-presidentialism has been made by Elgie
(1999, 2011) who not only clarified the concept but also realized a massive
program of empirical research on semi-presidentialism.
Albeit not without some controversy (Siaroff, 2003), the most widely
accepted definition of semi-presidentialism is the one proposed by Elgie
(1999, p. 13): “Semi-presidentialism is the situation where a constitution
b4407_Ch06.indd 136 1/6/2022 9:15:45 AM

makes provision for both a directly elected fixed-term president and a

prime minister and cabinet who are collectively responsible to the leg-
islature.” Therefore, semi-presidential systems are systems with directly
elected presidents who either share the power of ministerial appointment
with the legislature, which constitutes the president-parliamentary sub-
type, or possess other significant powers in the constitutionally defined
executive domains, which constitutes the premier-presidential subtype.
Semi-presidential systems used to be rather unusual, so that in his foun-
dational article, Duverger (1980) listed but five examples, complementing
the paradigmatic case of France with those of Austria, Finland, Iceland,
Ireland, and Portugal. At the time, these added cases were commonly
assigned to the category of parliamentarism because even if constitution-
ally strong, the presidents of these countries were not powerful in political
terms. Since then, however, semi-presidentialism became rather common
in democracies (Elgie, 2011).
The fourth type of executive–legislative relations was identified by
Shugart and Carey (1992, p. 26) as one in which the head of state is elected
not directly but rather by the parliament and yet endowed with significant
executive powers. Importantly, once elected, such a chief executive can-
not be removed by the assembly, which approximates the situation that
is typical of presidentialism. Citing just one example of such an institu-
tional arrangement, that of Switzerland, Shugart and Carey (1992, p. 26)
state that they “have little to add about this rare combination of powers.”
Other contemporary examples include the Federated States of Micronesia,
Suriname, and possibly San Marino. The term “assembly-independent
system” does not seem to be transparent semantically and, given that its
currency in the literature is limited because the phenomenon is very rarely
addressed, I prefer to use the term “systems with parliamentary elected
executive presidents.” The reason why this type of executive–legislative
relations is included into this study is that it in the past it had been adopted
by several authoritarian regimes, some of them presenting cases of great
significance.
b4407_Ch06.indd 137 1/6/2022 9:15:45 AM

6"×9"
The contemporary debate on the models of executive–legislative rela-

tions is largely centered around the question of these models’ suitability
for the sustainable development and efficient functioning of democracy.
Much of this debate is irrelevant to the study of authoritarianism, yet some
points deserve to be mentioned here. There is a long-standing argument,
most vocally exposed by Linz (1990), according to which parliamentarism
is more likely to lead to the successful consolidation of democracy than
presidentialism. Linz asserted that there was a linkage between this insti-
tutional arrangement and authoritarian trends in government, particularly
due to the perils of the presidential usurpation of power. The largely nor-
mative argument of Linz was challenged on empirical grounds (Power and
Gasiorowski, 1997; Ulfelder and Lustik, 2007). The approach of Shugart and
Carey (1992, p. 147) was more nuanced, as they concluded that while presi-
dential systems were not necessarily conducive to the failure of democracy
and even possessed some merits, powerful presidencies did pose a threat to
democracy. This argument has received support in more recent literature
that identifies executive hegemony as one of the major threats to emerging
democracies (Svolik, 2015; Pérez-Liñán et al., 2019).
One of the arguments of Linz (1990) explains the risks to democratic
stability posed by presidentialism with reference to the possibility of conflicts
between the executive and the legislature. Clearly enough, such risks are
even greater in semi-presidential systems where the government directly
depends on both sides of such conflicts, so that the cohabitation of the
mutually hostile presidents and assemblies can be particularly harmful.
Elgie (2008) recognizes the existence of a standard academic consensus that
semi-presidentialism is perilous for new democracies. But he also states that
this consensus is not sufficiently supported by the empirical evidence at hand,
which accords with the conclusions of more recent empirical studies (Cheibub
and Chernykh, 2009; Fernandes and Magalhães, 2016). Theoretically, how-
ever, there is no reason to deny that the risks of cohabitation are significant,
while the empirical evidence may be insufficient simply because the universal
spread of semi-presidentialism is a relatively recent phenomenon.
b4407_Ch06.indd 138 1/6/2022 9:15:45 AM

Executive–Legislative Relations and the

Properties of Authoritarian Party Systems
The task of assigning authoritarian party systems to the above models of
executive–legislative elections could be as difficult as any task involving a
thorough analysis of constitutional texts, but fortunately, it does not have
to be performed in this study due to the work of Elgie. Very comprehensive
lists of institutional arrangements by country, both historic and current,
are provided in some of his publications referred to above and consistently
updated at a specially developed website (Elgie, n.d.). Importantly, these
publications — in contrast to many others — contain information not only
about democracies but also about authoritarian regimes.
The results of my analysis, based mostly on these valuable sources but
also drawing some data from other publications (Ginsburg and Simpser,
2013), are presented in Table 6.1. Note that my approach is different from
that of Elgie in that he assigns to the category of monarchies only those
countries where the powers of the assemblies are extremely limited to
non-existent, such as Brunei or Oman, so that Morocco and Thailand fall into
the category of parliamentary systems. For a reason briefly explained above,
I count such countries as monarchies. The overall number of observations
listed in the table is greater than the number of authoritarian party systems
because some of the countries, as many as seven, changed their models of
executive–legislative relations but not their party systems. In such cases, years
of elections conducted when the given institutional arrangements were in
place are indicated in parentheses.
In order to proceed with this analysis, it is important to clarify the extent
to which the models of executive–legislative relations are contingent upon
authoritarian regime genomes, as defined in Chapter 5. The results of the
analysis are presented in Table 6.2. As is evident from the table, the associa-
tions are close enough but not sufficiently strong to allow for characterizing
institutional arrangements as mere products of political regimes. The most
direct correspondence can be observed between parliamentary systems and
b4407_Ch06.indd 139 1/6/2022 9:15:45 AM

6"×9"
Table 6.1. The classification of authoritarian party systems by

institutional types.
Political regime
types/origins
(numbers of
observations in
parentheses) Party systems
Monarchy (4) Iran, Morocco, Nepal, and Thailand
Parliamentary (10) Cambodia, Ethiopia, Guyana (1973), Malaysia,
Portugal, Sierra Leone (1967), Singapore, South Africa
(1948–1987), Zimbabwe (Rhodesia), and Zimbabwe
(1985)
System with Brazil, Guyana (1980–1985), Indonesia, and South
parliamentary elected Africa (1989)
executive president (4)
Presidential (31) Angola (1992), Bolivia, Burundi, Colombia, Republic
of Congo (2002–2012), Cote d’Ivoire, Djibouti,
Dominican Republic, Egypt (1976–2005), El Salvador
(1961–1978), El Salvador (1982–1991), Equatorial
Guinea, Gambia, Guatemala, Kazakhstan, Madagascar,
Mexico, Nicaragua, Paraguay, Philippines, Seychelles,
Sierra Leone (1973–1977), South Korea, South
Vietnam, Sudan, Taiwan, Tajikistan, Tunisia, Uganda,
Uzbekistan, and Zimbabwe (1990–2018)
Semi-presidential (15) Algeria, Angola (2008–2017), Armenia, Azerbaijan,
Cameroon, Chad, Republic of Congo (2017), Egypt
(2010), Gabon, Mauritania, Russia, Rwanda, Serbia,
Togo, and Yemen
party electoral authoritarian regimes. Systems with parliamentary elected

executive presidents are most likely to be found among military regimes
and post-democracies; presidentialism is characteristic of military regimes
and non-party electoral authoritarian regimes; and semi-presidentialism,
of single-party regimes and post-democracies.
Of course, the noticed affinities between institutional arrangements and
political regime genomes are directly related to these regimes’ properties.
Party electoral authoritarian regimes can only emerge in the conditions
when the party that is able to control a legislative majority is not only already
b4407_Ch06.indd 140 1/6/2022 9:15:45 AM

Table 6.2. Two-way analysis of variance: models of executive–legislative

relations and political regime genomes.
Other
Parliamentary systems, Pearson Cramér’s Fisher’s
systems, yes/no yes/no chi2 V exact
Military regime 0/63 57/174 19.28*** −0.26 0.00
Single-party regime 12/86 45/151 4.80** −0.13 0.03
Post-democracy 1/10 56/227 0.78 −0.05 0.70
Party electoral 44/36 13/201 89.19*** 0.55 0.00
Non-party electoral 0/21 57/216 5.44** −0.14 0.02
System with
parliamentary Other
elected executive systems,
presidents, yes/no yes/no
Military regime 11/52 3/228 28.51*** 0.31 0.00
Single-party regime 0/98 14/182 7.35*** −0.16 0.01
Post-democracy 2/9 12/271 4.54** 0.12 0.09
Party electoral 1/79 13/201 2.99* −0.10 0.12
Non-party electoral 0/21 14/259 1.13 −0.06 0.61
Other
Presidential systems,
systems, yes/no yes/no
Military regime 40/23 105/126 6.44** 0.15 0.02
Single-party regime 54/44 91/105 1.97 0.08 0.18
Post-democracy 0/11 145/138 11.12*** −0.19 0.00
Party electoral 35/45 110/104 1.36 −0.07 0.29
Non-party electoral 16/5 129/144 6.53*** 0.15 0.01
Other
Semi-presidential systems,
systems, yes/no yes/no
Military regime 12/51 45/186 0.01 −0.01 1.00
Single-party regime 32/66 25/171 16.55*** 0.24 0.00
Post-democracy 8/3 49/234 20.80*** 0.27 0.00
Party electoral 0/80 57/157 26.43*** −0.30 0.00
Non-party electoral 5/16 52/221 0.28 0.03 0.57
b4407_Ch06.indd 141 1/6/2022 9:15:46 AM

6"×9"
in place but also well entrenched due to its previous electoral record, as in
Malaysia and Singapore, or as a result of otherwise obtained superiority over
potential adversaries, as in Ethiopia. In contrast, military regimes normally
lack well-developed party structures and seek to consolidate their power
by endowing dictators with presidential powers, which can be achieved
under presidentialism or systems with parliamentary elected executive
presidents. External influences, sometimes conceptualized as institutional
diffusion (Brinks and Coppedge, 2006), also play a role. For example, it
has been noticed that in some of the Francophone countries of Africa, the
preference for semi-presidentialism was probably motivated by the experi-
ence of the former colonial power (Wu, 2011). In the post-Soviet countries
of South Caucasus and Central Asia, institution building was influenced
by the experience of Russia with its strong presidency operating within a
semi-presidential framework (Cameron and Orenstein, 2012).
This being said, it has to be re-emphasized that authoritarian institu-
tions arise largely from the endogenous choices of the autocrats. It is natural
that after coming to power under certain institutional arrangements, or
after creating them from scratch in the early periods of dictatorial rule, the
autocrats tend to keep them because they continue to serve their purposes
well. For the autocrat, frequent changes of the rules of the game are counter-
productive because they would not be consistent with one of the authoritar-
ian institutions’ key functions, alleviating commitment problems between the
dictator and his allies, and between the dictator and the permitted opposition
actors (Boix and Svolik, 2013). But if institutional change is deemed to be
necessary for regime survival, which is the case with numerous and mostly
successful attempts to extend the term limits of the incumbent presidents
(Reyntjens, 2016), then institutions can be reshuffled with a speed unthink-
able in democracies. Authoritarian institutions are endogenous primarily
because they can be changed for political gains at any time and with relative
ease. They remain stable as long as they deliver the desired results. This
brings us to the question of the autocrats’ incentive structure with regards
to party system parameters.
b4407_Ch06.indd 142 1/6/2022 9:15:46 AM

For the sake of theoretical parsimony, this structure of incentives can

be presented as comprising two essential goals. One of them is regime sur-
vival. The achievement of this goal can be threatened if the party system
is very fragmented and/or very volatile, and if parties with the primarily
territorial bases of support emerge as a threat to the national authorities.
The second goal is related to the fact that elections are a valuable component
of the authoritarian regimes that choose to conduct them. As discussed in
Chapter 1 of this book, elections facilitate power sharing and perform a
number of other functions ranging from substantively important to merely
symbolic. For these functions to be performed smoothly, elections have
to look credible, which implies that neither fragmentation nor volatility
of the party system have to be reduced to zero, and that the existence of
some parties with the primarily territorial bases of support can be tolerated.
In brief, this second goal can be defined as electoral credibility, which in
this context refers not to the normative standards of electoral governance
(Mozaffar and Schedler, 2002) but rather to the regime’s ability to ensure
that elections are viewed as competitive and fair by significant portions of
the elite and the population.
Obviously, this structure of incentives is internally contradictory. My
further theoretical reasoning is based on the assumption that the ways of
resolving the contradiction between the goals of regime survival and electoral
credibility vary in different institutional settings, and that this variation is
consequential for the parameters of authoritarian party systems. In par-
liamentary systems, the autocrat’s survival in power depends directly and,
speaking in institutional terms, exclusively on the ability of the pro-regime
party to win a legislative majority. This is consequential for party system
development in two respects. First, as noted in the recent literature (Roberts,
2015; Higashijima and Kasuya, 2016), the leaders of parliamentary author-
itarian regimes have to pay special attention to the maintenance of the
dominant party, which involves high degrees of organizational coherence
and intra-party power sharing. In order to function as a reliable instru-
ment of autocracy, the dominant party has to continuously display both
b4407_Ch06.indd 143 1/6/2022 9:15:46 AM

6"×9"
organizational stability and very high levels of electoral s upport. In these

conditions, the goals related to electoral credibility are still important but
secondary.
The importance of the dominant party’s cohesion and continuous
electoral superiority has been demonstrated by a largely unanticipated
transition to electoral democracy in Malaysia that occurred as a result of the
2018 elections (Nadzri, 2018). For several decades after independence, the
country’s party system remained highly concentrated as a result of the con-
tinuous domination of the ruling coalition, the National Front, and its core
component party, the United Malays National Organization. Starting from
2008, the electoral standing of the previously insignificant and fragmented
opposition gradually improved, so that the party system became more frag-
mented and more volatile than previously. But the decisive defeat of the ruling
coalition in the 2018 elections occurred only after a veteran National Front
politician and the former prime minister, Mahathir Mohamad, defected and
founded his own party, the Malaysian United Indigenous Party, that joined
the coalition of the previously existing opposition groups (Lemière, 2018).
The example of Malaysia, by combining the emergence of viable
opposition parties with the breaking up of cohesion in the dominant
party, illustrates the worst-case scenario in the trajectory of a parliamen-
tary authoritarian regime. Note that the two aspects are interconnected.
The cohesion of the ruling party is quite unlikely to be broken up in the
conditions when opposition parties are so weak that joining them cannot
be viewed as a path toward electoral success. Hence, we can expect that in
order to avoid trouble, parliamentary authoritarian regimes will fully employ
their repertoire of manipulation, as described in Chapter 1, to keep both
fragmentation and volatility of their party systems at the lowest possible
level. Even high fragmentation within the opposition camp is hardly toler-
able for the parliamentary-based autocracies because in order to limit the
scope of the opposition parties’ appeal to different societal constituencies,
the number of these parties has to be restricted. This has clear implications
for party system format. At the same time, parties with primarily territorial
b4407_Ch06.indd 144 1/6/2022 9:15:46 AM

bases of support can be tolerated, but only on the condition that these bases
exist in very small territorial units, so that the overall level of party system
nationalization remains rather high.
Is this argument valid for systems with parliamentary elected execu-
tive presidents? At first glance, it should be. These systems are similar to
parliamentarism in the sense that under both institutional arrangements,
winning parliamentary elections is a necessary prerequisite for staying in
power. There is a very illustrative example of a long-standing authoritarian
regime that collapsed in a constellation of circumstances that shared cer-
tain characteristics with the case of Malaysia. During the long period of
its existence, the military dictatorship of Brazil employed the system with
parliamentary elected executive presidents. In 1985, a large faction favor-
ing direct presidential elections broke away from the pro-regime National
Renewal Alliance and launched a new party, the Liberal Front, later to join
forces with the anti-regime opposition (Mainwaring, 1986). Soon after the
opposition coalition won a majority of seats in the national legislature, one
of the leaders of the Liberal Front, José Sarney, assumed the presidency
and presided over Brazil’s transition to democracy. One would argue that
the Brazilian dictatorship would be better off, at least in terms of avoiding
electoral threats, with a more cohesive ruling party and a weaker opposition,
which was indeed the case in the other historically prominent system with a
parliamentary elected executive president, Indonesia under Suharto. Briefly
before the regime of Suharto fell under the pressure of popular unrest and
military defection, the ruling Golkar party was still able to win elections by
a huge margin (Eklöf, 1997).
In fact, a juxtaposition of the cases of Brazil and Indonesia reveals that
the incentive structures of systems with parliamentary elected executive
presidents are different from those under parliamentarism. Genetically,
both Brazil and Indonesia were military regimes. During the formative
phases of the dictatorships, institutionalizing them in the form of parlia-
mentary systems was not a feasible option. But, once the leaders of these
regimes decided to avoid the risks associated with direct presidential races
b4407_Ch06.indd 145 1/6/2022 9:15:46 AM

6"×9"
by adopting systems with parliamentary elected executive presidents, the

lack of inherited pro-regime parties remained a problem. Both countries
resolved this problem by creating controlled party systems with limited
numbers of permitted opposition parties. At least originally, these parties
were conceived not as a threat to the survival of the authoritarian regimes
but rather as their useful assets. Thus, in both cases, it is possible to register
incentives for keeping fragmentation at a low but not negligible level.
Similar incentives, albeit of a very different origin, can be registered in
Guyana and South Africa that adopted systems with parliamentary elected
executive presidents after abandoning parliamentarism. When launching
constitutional reforms, the leaders of these regimes sought to enhance
their leverage over the dominant parties by cutting off their influence in
the day-to-day activities of the national government. Such an incentive
structure is particularly apparent in the case of South Africa, where the
executive presidency was introduced specifically with the aim of increasing
the strength of the executive vis-à-vis the dominant party in the wake of a
political transformation that could be painful for the party’s support base
(Spence, 1991, p. 325).
Thus, there is no reason to expect that the system with parliamentary
elected executive president will reduce fragmentation to the same extent
as the parliamentary system. At the same time, the examples of Brazil and
Indonesia indicate that the leaders of the regimes, being aware of the risks
to regime survival associated with a more permissive attitude toward the
acceptable degrees of electoral success of the opposition, sought to minimize
these risks by controlling party system volatility, which was achieved by
restricting entry to the electoral arena. As a side effect, this can lead to the
exclusion of parties with primarily territorial bases of support, resulting in
high party system nationalization.
In presidential systems, the survival of the executive does not depend on
the political composition of the assembly. Thus, the risks to regime survival
stemming from party system properties are much lower than is the case with
parliamentary systems, particularly because the leader of the regime can
b4407_Ch06.indd 146 1/6/2022 9:15:46 AM

run in elections without any party affiliation. Among the large number of
presidential systems included in this study, there is a clear-cut example of an
overtly presidential regime where the effective number of parties has always
been set at the level comparable with stable European liberal democracies,
Uzbekistan. The presidents of the country, first Islam Karimov and then
Shavkat Mirziyoyev, did not head any political parties and won their carefully
staged elections as independents, but the arena of parliamentary elections
was open only to parties that unequivocally supported the executives and
their self-styled “national ideology” (Omelicheva, 2016). Moreover, the
party system of Uzbekistan had been periodically reshuffled, with some of
the parties falling out of favor with the president, as a result of which they
were replaced by more suitable newcomers.
Of course, Uzbekistan stands out as an example of an extremely
restrictive, overtly dictatorial political regime. But several Latin American
countries, including El Salvador and Guatemala, may serve as examples
of multiparty politics in more permissive political environments of pres-
idential authoritarianism. In these countries, the incumbent leaders were
normally party-affiliated, and they often confronted significant risks at
the polls. At times, this even led to alternation in power, even though the
circle of potential winners was limited to the narrow ruling groups of these
military-controlled regimes. Some of these regimes score extremely low on
the parameter of party system fragmentation, as exemplified by Djibouti
where the ruling People’s Rally for Progress continuously gained all seats
in the legislature in 1992–2008. In general, however, there are reasons to
expect that the independence of the executive from the legislature, both in
terms of their origin and survival, makes authoritarian presidential systems
more conducive to fragmented and volatile patterns of party politics than
it is the case with the two previously discussed institutional arrangements.
When investigating the interplay of presidentialism and party system
fragmentation in democracies, several scholars have advanced a theory
according to which presidentialism shapes legislative fragmentation pri-
marily by exerting coattail effects, with a party’s prospects for success at the
b4407_Ch06.indd 147 1/6/2022 9:15:46 AM

6"×9"
legislative level being contingent upon this party’s prospects in the presiden-
tial election (Jones, 1994; Shugart, 1995). Several cross-national empirical
studies have concurred with the hypothesis that in this way, presidentialism
reduces party system fragmentation (Hicken and Stoll, 2008; Williams-
Wyche, 2014), while other studies go even further by demonstrating that even
without the coattail effects, the effective number of parties in presidential
systems tend to be relatively low (Golosov and Kalinin, 2017). While there
is no reason to argue for the equivalence of these effects in democracies and
autocracies in terms of party system fragmentation, I expect that the coattail
effects can affect the performance of opposition parties, thereby reducing
their number to those that can run visible candidates in presidential races.
This can affect the party system format by increasing the probability of
dominant party system with low opposition fragmentation.
There is an argument according to which in democracies presidential-
ism is positively related to party system nationalization because presidential
elections naturally suppress local sentiment in the electorate by appealing to
national political forces (Shugart and Carey, 1992; Hicken and Stoll, 2011).
Other studies either reject this theory on theoretical (Morgenstern et al.,
2009) or empirical (Golosov, 2016a) grounds, or accept it only with qualifica-
tions, noting, in particular, that “presidential elections with few presidential
candidates promote the nationalization and consolidation of the legislative
party system only when the president is neither very weak nor very pow-
erful” (Hicken and Stoll, 2013, p. 291; see also Stoll, 2015), which suggests
that the argument does not apply to authoritarian regimes. However, there
is certainly no reason to expect that the level of party system nationalization
under authoritarian presidential regimes will be low in comparison to other
varieties of non-democratic institutional arrangements.
The incentive structure of authoritarian semi-presidentialism is
particularly complex. Under this institutional arrangement, the regime’s
hold on power can be fully sustained only by meeting two conditions: that
the incumbent executive wins presidential elections, and that the main
pro-regime party gains a legislative majority. Does it mean that the effects
b4407_Ch06.indd 148 1/6/2022 9:15:46 AM

of semi-presidentialism upon party system properties will be similar in

direction, and probably even greater in scope, than those of parliamen-
tarism? Prior to answering this question, it is important to address the
implicit assumption that the election-related risks to regime survival under
semi-presidentialism are great. Yet in fact, the share of semi-presidential
systems among the contemporary dictatorships is so large that these systems’
wide spread cannot be explained solely with reference to diffusion, particu-
larly given that the autocrats can easily abandon inconvenient institutional
arrangements. Rather, it indicates that the autocrats themselves not only view
the system as relatively risk-free but also find it advantageous for certain
reasons, some of which can be related to risk avoidance.
One of the most turbulent phases in the history of any personalist
dictatorship with a directly elected president is the moment when the consti-
tutionally established presidential term limit expires, so that the incumbent
president loses the entitlement to hold office. Of course, pure presidential-
ism endows the incumbent with only two choices, to quit or to change the
constitution. The latter choice is quite popular among the dictatorial exec-
utives, but it can entail a serious political crisis (Baturo and Elgie, 2019).
Semi-presidentialism allows for avoiding such complications.
In 2008, the incumbent president of Russia, Vladimir Putin, solved the
problem of term limits by allowing his close political ally, prime minister
Dmitry Medvedev, to run in presidential elections. At that time, Medvedev
lacked personal popularity or any other significant political assets, so it was
clear enough that he could win only if endorsed by Putin as his successor.
Putin himself headed the election list of the main pro-regime party, United
Russia, in the national legislative elections held several months before the
presidential ones. After Medvedev did indeed assume the presidency in 2008,
Putin became the prime minister and, in effect, continued to run the country
(White and McAllister, 2008). In 2012, when Putin was entitled to claim the
presidency again, he ran and won. This time, Putin did not head the list of
United Russia in the preceding legislative elections and stood as its candi-
date only in the presidential races, with his candidacy being also endorsed
b4407_Ch06.indd 149 1/6/2022 9:15:46 AM

6"×9"
by two other registered parties and a large number of public associations

that formed an entity called “All-Russia People’s Front” (Aburamoto, 2019;
Tkacheva and Golosov, 2019).
The party system of the Democratic Republic of the Congo is not
included in this study because of the extremely high level of volatility, but
the political dynamics of semi-presidentialism in this country are illustrative
enough to be addressed here. The incumbent president, Joseph Kabila, failed
in his numerous attempts to create the constitutional conditions for the third
term in office. In the 2018 presidential elections, Kabila supported one of
the candidates, Félix Tshisekedi, who differed from Russia’s Medvedev in
that he did have some opposition credentials and resources of his own, but
these were insufficient for defeating other contenders without the support
of the outgoing president. Thus, Tshisekedi owed his victory in presidential
elections to Kabila’s support. But more than this, within the framework of
semi-presidentialism, Tshisekedi could only form a government with the
support of Kabila’s People’s Party for Reconstruction and Democracy that,
jointly with its regional allies, won a solid parliamentary majority. Thus,
Kabila effectively retained a large share of power (Berwouts and Reyntjens,
2019).
The examples above show that semi-presidentialism endows the incum-
bent autocrats with a greater ability to avoid risks associated with directly
elected presidency, which, putting it in the language of business, can be
referred to as risk hedging. Note that under this institutional arrangement,
the reduction of risks does not necessarily entail severe restrictions on the
activities of opposition parties. Neither Putin nor Kabila would be better
off with fewer opposition parties in parliamentary races, and the example of
the Democratic Republic of the Congo is particularly illustrative in showing
that high volatility is not very problematic either. At the same time, once
the risks to regime survival are minimized, it becomes possible to maximize
gains in terms of electoral credibility. Thus, it can be expected that the party
systems of semi-presidential systems will be relatively volatile, fragmented,
and, with respect to party system format, they will allow for party dominance
b4407_Ch06.indd 150 1/6/2022 9:15:46 AM

with high fragmentation within the opposition camp. There is no reason to

expect that parties with territorial bases of support will be excluded.
The above-stated expectations can be summarized in the form of the
following hypotheses:
H1. Under parliamentarism, the authoritarian party systems will be

characterized by low fragmentation with some propensity toward
the party system format with low opposition fragmentation. These
party systems will also score low on the parameters of volatility and
nationalization.
H2. Under the systems with parliamentary elected executive presidents,
authoritarian party systems will be generally similar to those of
parliamentary regimes, but they will allow for greater degrees of
fragmentation. High levels of party system nationalization can be
expected.
H3. Under presidentialism, authoritarian party systems will be char-
acterized by moderate scores on all parameters, thus exceeding
the systems with parliamentary elected executive presidents
on the parameter of fragmentation, and they will have some
propensity toward the party system format with low opposition
fragmentation.
H4. Under semi-presidentialism, the authoritarian party systems will
be characterized by relatively high fragmentation and volatility,
and they will have some propensity toward the party system format
with high opposition fragmentation.
When testing these hypotheses, I follow the same approach as in

Chapter 5 by creating combined categories. Consistent with the above-stated
hypotheses, I define these categories as “Systems with parliamentary elected
executive presidents + Presidential systems” in the case of party system
fragmentation, and “Parliamentary systems + Systems with parliamentary
elected executive presidents” in the case of party system volatility.
b4407_Ch06.indd 151 1/6/2022 9:15:46 AM

6"×9"
The empirical findings presented in Tables 6.3–6.7 pertain to the effects

of the modes of executive–legislative elections upon party system fragmen-
tation. In accordance with the above hypotheses, the analysis demonstrates
that the party systems of authoritarian parliamentary regimes are indeed
significantly less fragmented than those of the systems with parliamentary
elected executive presidents, presidential, and semi-presidential systems. At
the same time, the difference between semi-presidentialism and the com-
bined category is insignificant, even though the direction of the observed
association concurs with H4. Table 6.6 shows that among all institutional
arrangements, only parliamentarism exerts a significant negative impact
upon party system fragmentation. However, all coefficients in the multilevel
Table 6.3. Party system fragmentation by models of executive–legislative

relations.
Standard
Parliamentary systems 1.29 1.15 1.00 2.31 0.34 57
Systems with parliamentary 1.58 1.51 1.29 2.20 0.27 14
elected executive
presidents
Presidential systems 1.51 1.36 1.00 4.28 0.59 145
Semi-presidential systems 1.83 1.52 1.00 4.59 0.81 57
elected executive
presidents + presidential
systems
Table 6.4. Summary of ANOVA comparison of institutional design types

by fragmentation.
Between groups 209.63 2 104.82 101.61 0.00
Within groups 300.19 291 1.03 — —
Total 509.83 293 — — —
b4407_Ch06.indd 152 1/6/2022 9:15:46 AM

Table 6.5. Bonferroni corrected comparisons of institutional design

types by fragmentation.
Mean Standard
Monarchy Parliamentary 3.47 0.26 0.00
System with parliamentary 3.25 0.24 0.00
elected executive
president + presidential
Semi-presidential 2.93 0.26 0.00
Parliamentary Monarchy −3.47 0.26 0.00
System with parliamentary −0.22 0.16 0.97
elected executive
Semi-presidential −0.54 0.19 0.03
System with Monarchy −3.25 0.24 0.00
parliamentary Parliamentary 0.22 0.16 0.97
elected executive
president + Semi-presidential −0.32 0.16 0.26
presidential
Semi-presidential Monarchy −2.93 0.26 0.00
Parliamentary 0.54 0.19 0.03
System with parliamentary 0.32 0.16 0.26
elected executive
model, as reported in Table 6.7, are significant, and their values are ordered
in the predicted way, so that semi-presidential systems are associated with
most fragmented party systems, followed by the systems with parliamentary
elected executive presidents and presidential systems, and distantly traced by
parliamentarism. Note that the combined category displays medium scores
in all models, which is consistent with H2.
Figure 6.1 and Table 6.8 clarify the findings reported above with regards
to party system format. Consistent with the hypotheses, parliamentary and
presidential systems are disposed toward predominant party systems with
concentrated opposition, while semi-presidentialism is associated with
high fragmentation within the opposition camp. There are no significant
b4407_Ch06.indd 153 1/6/2022 9:15:46 AM

6"×9"
Table 6.6. Linear regression results, effective number of parties by

institutional design types (n = 294 adjusted for 57 clusters; unstandardized
coefficients; clustered-robust errors in parentheses).
System with
parliamentary
elected executive
president +
Parliamentary Presidential Semi-presidential
Institutional design types −0.57** (0.23) −0.56 (0.39) 0.01 (0.29)
Index of electoral 2.25* (1.16) 1.37 (1.25) 2.17 (1.37)
democracy
Index of ethnic −0.09 (0.30) −0.17 (0.31) −0.14 (0.28)
fractionalization
Second elections −0.23** (0.11) −0.16* (0.09) −0.19** (0.10)
Real GDP per capita in −0.01** (0.01) −0.02** (0.01) −0.02** (0.01)
2011 USD equivalent
(thousands)
Constant 1.41*** (0.33) 1.88*** (0.58) 1.36*** (0.34)
R-squared 0.05 0.07 0.03
Table 6.7. Multilevel mixed-effects regression results, effective number

of parties by models of executive–legislative relations.
Standard
Parliamentary −2.87 0.49 0.000
System with parliamentary elected −2.74 0.45 0.000
executive president + presidential
Index of ethnic fractionalization −0.58 0.45 0.197
Constant 4.21 0.55 0.000
Country 0.80 0.09 —
Year 0.00 0.00 —
Note: Wald chi2 = 45.51; Log restricted-likelihood = −348.35; Prob > chi2 = 0.00.
b4407_Ch06.indd 154 1/6/2022 9:15:46 AM

1 1
Plate 1 Plate 2
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 1
Plate 3 Plate 4
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

in
authoritarian elections by institutional design types. Plate 1:
Parliamentary systems. Plate 2: Systems with parliamentary elected
executive presidents. Plate 3: Presidential systems. Plate 4: Semi-
presidential systems.
associations between the systems with parliamentary elected executive

presidents and any of the party formats. It is also noticeable that unlike all
other types of executive–legislative relations, semi-presidentialism is not
negatively associated with party system fragmentation, even though the
positive association is statistically insignificant.
Tables 6.9–6.13 show the results of the analysis with respect to the insti-
tutional effects upon party system volatility. As is evident from the tables,
the party systems belonging to the combined category of parliamentary
systems and systems with parliamentary elected executive presidents are
more stable than those in all other categories, which is consistent with my
hypotheses. The semi-presidential authoritarian party systems demonstrate
b4407_Ch06.indd 155 1/6/2022 9:15:46 AM

6"×9"
Table 6.8. Two-way analysis of variance: models of executive–legislative

relations and party system formats.
Predominant
party system Other
with fragmented systems, Pearson Cramér’s Fisher’s
opposition, n n chi2 V exact
Parliamentary 22 35 0.02 0.01 0.88
System with 4 10 0.53 −0.04 0.58
parliamentary elected
executive presidents
Presidential 51 94 0.81 −0.05 0.40
Semi-presidential 33 24 12.20*** 0.20 0.00
Predominant
party system with Other
concentrated systems,
opposition, n n
Parliamentary 29 28 4.90** 0.13 0.03
System with 7 7 0.88 0.05 0.40
Presidential 65 80 5.50** 0.14 0.02
Semi-presidential 8 49 17.36*** −0.24 0.00
Other
Systems without systems,
majority party, n n
Parliamentary 6 51 7.166*** −0.16 0.01
System with 3 11 0.06 −0.01 1.00
Presidential 29 116 2.69* −0.10 0.11
Semi-presidential 16 41 0.59 0.04 0.49
at 0.1 level.
a rather high level of volatility. In this respect, however, they are not signifi-
cantly different from the presidential authoritarian party systems. Table 6.12
shows that among all institutional influences, only the combined category
of parliamentary systems and systems with parliamentary elected executive
b4407_Ch06.indd 156 1/6/2022 9:15:46 AM

Table 6.9. Party system volatility by models of executive–legislative

relations.
Standard
elected executive presidents
Parliamentary systems + 9.49 5.39 0.00 44.72 10.41 59
systems with parliamentary
elected executive presidents
Table 6.10. Summary of ANOVA comparison of institutional design

types by volatility.
Between groups 0.40 3 0.13 7.66 0.00
Within groups 4.09 233 0.02 — —
Total 4.49 236 — — —
presidents serves as a significant — and negative — predictor of party system

volatility. But the directions of the coefficients are positive, as expected, for
the two other variables pertaining to institutional design. Unlike in the above
analysis of party system fragmentation, the multilevel model, as reported in
Table 6.13, fully concurs with this result in the sense that only the impact of
the combined category is significant. However, the model ranks the values
of the coefficients in the theoretically expected way.
The empirical findings presented in Tables 6.14 and 6.15 pertain
to the effects of the modes of executive–legislative relations upon party
system nationalization. The findings, however scarce, fully confirm
H2 by showing that the systems with parliamentary elected executive
presidents are indeed associated with the highest levels of party system
b4407_Ch06.indd 157 1/6/2022 9:15:46 AM

6"×9"
Table 6.11. Bonferroni corrected comparisons of institutional design

types by volatility.
Mean Standard
Monarchy Parliamentary + system 0.15 0.04 0.00
with parliamentary elected
executive president
Presidential 0.09 0.03 0.04
Parliamentary + Monarchy −0.15 0.04 0.00
system with Presidential −0.06 0.02 0.03
parliamentary elected Semi-presidential −0.09 0.03 0.00
executive president
Presidential Monarchy −0.09 0.03 0.04
Parliamentary + system 0.06 0.02 0.03
executive president
Semi-presidential Monarchy −0.06 0.04 0.64
Parliamentary + system 0.09 0.03 0.00
executive president
Table 6.12. Linear regression results, volatility by institutional design

types (n = 294 adjusted for 57 clusters; unstandardized coefficients;
Parliamentary + system
with parliamentary
elected executive
president Presidential Semi-presidential
Institutional design −0.06** (0.03) 0.004 (0.027) 0.03 (0.03)
types
Index of electoral 0.35** (0.15) 0.39** (0.17) 0.35* (0.18)
democracy
Index of ethnic 0.03 (0.04) 0.02 (0.04) 0.02 (0.04)
fractionalization
Second elections 0.06*** (0.02) 0.06*** (0.02) 0.06*** (0.02)
Real GDP per capita in −0.002* (0.001) −0.003* (0.001) −0.003* (0.001)
2011 USD equivalent
(thousands)
Constant 0.06 (0.04) 0.05 (0.05) 0.06 (0.05)
R-squared 0.14 0.11 0.11
b4407_Ch06.indd 158 1/6/2022 9:15:46 AM


models of executive–legislative relations.
Standard
Parliamentary + system with −0.13 0.05 0.018

parliamentary elected executive
president
Presidential −0.06 0.05 0.205
Index of ethnic fractionalization 0.02 0.05 0.679
Constant 0.15 0.06 0.023
Country 0.07 0.01 —
Year 0.07 0.02 —
Table 6.14. Party system nationalization by models of executive–

legislative relations.
Standard
Systems with 0.86 0.86 0.83 0.89 0.04 2
Table 6.15. Pairwise correlations (Pearson’s), nationalization by

institutional design types, n = 42.
chi2 Significance
Parliamentary 0.18 0.27
Systems with parliamentary elected executive president 0.18 0.25
Presidential 0.27 0.09
Semi-presidential –0.32 0.04
b4407_Ch06.indd 159 1/6/2022 9:15:46 AM

6"×9"
nationalization, and that the level of nationalization of the authoritarian

semi-presidential party systems is low as expected. Both associations are
statistically significant.
Federalism and the Properties of

The question of whether authoritarian federalism is possible at all is con-
troversial. Riker (1975, p. 101) defined federalism as “a political organi-
zation in which the activities of government are divided between regional
governments and a central government in such a way that each kind of
government has some activities on which it makes final decisions,” which
apparently suggests that authoritarianism, since it does not necessarily
prohibit such a division of activities, may be consistent with federalism.
As argued by Diamond (1999, p. 150), the definition of Riker still suggests
that the autonomy of sub-national units should be constitutionally guar-
anteed. Since such guarantees are absent under the presumably arbitrary
rule of autocrats, authoritarian federalism is an oxymoron, which is a view
exposed by Diamond and shared by many other leading scholars in the
field (Elazar, 1991; Stepan, 2001). While radically dissenting theoretical
views are rare (Kahn, 2002), there is an increasing number of academic
publications in which the phenomenon of authoritarian federalism is dealt
with in a more nuanced way. For instance, Barracca (2007) argues that
the basic criterion set by Riker, the vertical division of power between the
central and sub-national authorities, can be met by authoritarian regimes,
but the rules of the federal pact are especially prone to violation in the
absence of democracy.
In essence, of course, this debate pertains to the wider issue of the
consequentiality of endogenous institutions. Clearly, both democratic and
authoritarian rulers can violate the constitutionally defined rules of the
game for political gains, and it is also clear that under authoritarianism, the
b4407_Ch06.indd 160 1/6/2022 9:15:47 AM

probability of such violations is particularly high. To the same extent as with

other institutional arrangements, this is the case with federalism. However,
the autocrats set these rules not in order to violate them, for which end they
are useless, but rather in order to follow them as long as they do not pose an
obvious and/or immediate threat to their interests. This implies that author-
itarian federalism is different from the “true” federalism of democracies, on
the one hand, and that it still can be consequential, on the other hand. But
there is no reason to expect that the effects of federalism in authoritarian
and democratic settings will be the same.
This study includes six authoritarian party systems that emerged
and/or functioned under federal constitutions: Brazil, Ethiopia, Malaysia,
Mexico, and Russia, where such constitutions were in place throughout
the whole observed periods, and Sudan, where the elections of 2010 and
2015 were held in the conditions of federalism. Among those listed, the
cases of long-standing federalism attracted careful scholarly attention.
The literature clearly suggests that in all these cases, federalism was
not inconsequential, even though the observed effects of authoritarian
federalism in different domains of policy and politics were sometimes
idiosyncratic and varied in different national contexts, as evident from the
studies of Brazil (Samuels and Abrucio, 2000), Ethiopia (Abbink, 2011),
Malaysia (Case, 2007), Mexico (Ward and Rodriguez, 1999), and Russia
(Starodubtsev, 2018).
In this book, the effects of federalism have to be estimated only from
the narrow perspective of its impact upon the properties of authoritarian
party systems. As any kind of division of power, federalism poses poten-
tial threats to authoritarianism, and there is a reason to expect that these
threats can materialize in the electoral arena. Elections, while leading to
important national political outcomes, are conducted locally, so that the
authorities of the federal units, if endowed with significant powers under
federal arrangements, can influence the composition of national legislature.
Within the resulting incentive structure, it is imperative for the national
b4407_Ch06.indd 161 1/6/2022 9:15:47 AM

6"×9"
political executive to ensure that the sub-national votes, rather than being
absorbed by a plethora of parties controlled by the regional authorities, will
be delivered to those parties that directly serve the interests of the national
autocratic government. At the same time, high levels of autonomy enjoyed
by federal units, especially if combined with the societal foundations of party
support such as ethnicity, can contribute to the sustainability of regional
parties. Several studies have found that in democracies, federalism is nega-
tively related to party system nationalization (Jones and Mainwaring, 2003;
Chhibber and Kollman, 2004; Golosov, 2016a; see however Thorlakson,
2007; Morgenstern et al., 2009), which may have implications for other
party system properties.
With respect to party system formation, the observed structure of
incentives can produce varying outcomes that are contingent upon the
country-specific political trajectories. Russia’s party system, as it emerged
and started to develop under the conditions of imperfect democracy in the
1990s, was very fragmented and displayed a low level of nationalization
(Golosov, 2015c). By the end of the 1990s, national election outcomes
were heavily influenced by regional authorities and, primarily, by the
chief executives of the regions who combined vast political resources with
strong institutional standing vis-à-vis all other political actors, national
and sub-national alike (Demchenko and Golosov, 2016; Golosov and
Konstantinova, 2016). In the process of the creation of the main pro-regime
party, United Russia, regional political machines were largely absorbed
by it, while the emergence of new parties with the regional bases of sup-
port was effectively arrested by restrictive legislation on political parties
(Golosov, 2014c). By the end of the 2010s, Russia’s party system was highly
concentrated and nationalized. It can be noticed that in the authoritarian
party system of Mexico, some of the opposition parties did have territorial
bases of support (Loaeza, 2000), but the main pro-regime party was highly
nationalized, which resulted in a concentrated and nationalized party system
much as it happened in Russia.
A different pattern of party system development can be observed in
Ethiopia where the development of nationally dominant parties did not
b4407_Ch06.indd 162 1/6/2022 9:15:47 AM

entirely eliminate political parties with the territorial bases of support.

The ruling Ethiopian People’s Revolutionary Democratic Front emerged
as a coalition of the strongest military-political groups that succeeded
in their fight against the previous dictatorial regime, but, for a variety of
reasons, some of the ethno-regional parties remained outside of the rul-
ing party and continued to run their regions on the condition that they
did not contest the supremacy of the national leadership (Kefale, 2011).
These parties were present in the national elections as well, which led to
a relatively low degree of party system nationalization. In Malaysia, the
ruling National Front generally followed the pattern observed in Russia
by absorbing most of the regional elites. But in some parts of the country,
particularly in Sarawak and Sabah, regional parties managed to survive,
which was tolerated, even if not very willingly, by the National Front.
At times, regional parties participated in the regional governments and
achieved modest yet visible representation in the national parliament
(Hazis, 2018).
This internally contradictory structure of incentives for party system
development does not allow for formulating any straightforward theoretical
expectations regarding the impact of authoritarian federalism on party sys-
tem fragmentation and volatility. But the relative weight of the two structures
of incentives can be established empirically. With regards to party system
format, however, there is a solid reason to expect that if opposition parties
are tolerated to any visible extent, then:
H5. Authoritarian federalism will yield dominant party systems with

fragmented opposition.
Tables 6.16–6.18 show that the party systems of authoritarian federations

are less fragmented and volatile than the party systems of non-federal states.
On both parameters, however, the statistical significance of the observed
differences is weaker than at the 0.05 level. Note that for Tables 6.17 and 6.18,
no post hoc tests were performed because there are fewer than three groups.
With respect to party system nationalization, no difference whatsoever can be
b4407_Ch06.indd 163 1/6/2022 9:15:47 AM

6"×9"
Table 6.16. Party system parameters in relation to federalism.

Standard
Fragmentation, 1.42 1.35 1.04 2.31 0.35 43
federations
Fragmentation, 1.84 1.40 1.00 11.80 1.43 243
non-federations
Volatility, federations 0.12 0.10 0.00 0.33 0.10 38
Volatility, 0.16 0.13 0.00 0.65 0.14 199
non-federations
Nationalization, 0.71 0.72 0.55 0.89 0.14 6
federations
Nationalization, 0.71 0.72 0.29 0.98 0.19 36
non-federations

non-federations by fragmentation.
Between groups 6.18 1 6.18 3.58 0.06
Within groups 503.65 292 1.72 — —
Total 509.83 293 — — —

non-federations by volatility.
Between groups 0.05 1 0.05 2.74 0.10
Within groups 4.44 235 0.02 — —
Total 4.49 236 — — —
found between federations and non-federations. This makes it unnecessary

to perform further statistical tests on this parameter.
Tables 6.19 and 6.20 clarify these findings by assessing the impact
of authoritarian federalism upon party system properties. The linear
b4407_Ch06.indd 164 1/6/2022 9:15:47 AM

Table 6.19. Linear regression results, fragmentation and volatility by

federalism (n = 294 adjusted for 57 clusters; unstandardized coefficients;
Fragmentation Volatility
Federalism −0.71* (0.40) −0.04 (0.04)
Index of electoral democracy 2.38* (1.28) 0.40** (0.18)
Index of ethnic fractionalization −0.03 (0.33) 0.03 (0.04)
Second elections −0.21** (0.10) 0.06*** (0.02)
Real GDP per capita in 2011 USD −0.01* (0.01) −0.003* (0.001)
Constant 1.28*** (0.35) 0.05 (0.05)
R-squared 0.05 0.12
at 0.1 level.
Table 6.20. Multilevel mixed-effects regression results, fragmentation

and volatility by federalism.
Fragmentation Volatility
Federalism −0.29 (0.45) −0.04 (0.05)
Index of electoral democracy 1.74 (1.01)* 0.31 (0.16)*
Index of ethnic fractionalization −0.23 (0.58) 0.02 (0.05)
Second elections −0.18 (0.10)* 0.04 (0.02)**
Real GDP per capita in 2011 USD −0.01 (0.01) −0.001 (0.001)
Constant 1.50 (0.39)*** 0.07 (0.05)
Country 1.06 (0.11) 0.07 (0.01)
Year 0.000 (0.001) 0.07 (0.01)
Notes: Fragmentation model: Wald chi2 = 7.28; Log restricted-likelihood = –366.21;

Prob > chi2 = 0.30.
Volatility model: Wald chi2 = 12.11; Log restricted-likelihood = 140.66; Prob >
chi2 = 0.06.
*** significant at 0.01 level, ** significant at 0.05 level, and * significant at
0.1 level.
b4407_Ch06.indd 165 1/6/2022 9:15:47 AM

6"×9"
regression model does confirm that authoritarian federalism is negatively

related to party system fragmentation, but this is not supported by the
multilevel regression results. The impact of authoritarian federalism
upon party system volatility is insignificant in both models. Note,
1 1
Plate 1 Plate 2
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

federal and non-federal autocracies. Plate 1: Federations. Plate 2:
Non-federations.
Table 6.21. Two-way analysis of variance, federalism, and party system

formats.
Predominant
party system
with fragmented Other Pearson Cramér’s Fisher’s
opposition, n systems, n chi2 V exact
Federations 15 28 16.05*** 0.23 0.00
Predominant
party system with
concentrated Other
opposition, n systems, n
Federations 11 32 3.34* −0.11 0.09
Systems without Other
majority parties, n systems, n
Federations 4 39 6.06** −0.14 0.01
at 0.1 level.
b4407_Ch06.indd 166 1/6/2022 9:15:47 AM

however, that the signs of the related coefficients are invariably negative.
At the same time, the data in Figure 6.2 and Table 6.21 fully confirm
H5 not only by showing that authoritarian federalism does indeed yield
predominant party systems with fragmented opposition, but also by
demonstrating that other party system formats, and especially systems
without majority parties, are unlikely to be found under this institutional
arrangement.
b4407_Ch06.indd 167 1/6/2022 9:15:47 AM

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Chapter 7

and Electoral Systems
E
lectoral system is a multifaceted concept that can refer to many
aspects of the electoral process, including the rules of party
registration, voter registration, political finance, media coverage of
elections, and so forth. In the analysis that follows, I deal exclusively with
the aspects that are central for the operation of any electoral system, iden-
tified by Lijphart (1990, p. 481) as the triad of electoral formula, district
magnitude, commonly understood as the number of representatives elected
in the electoral districts, and ballot structure, commonly divided into three
main varieties: closed, flexible, and open (Däubler and Hix, 2018). The first
element, electoral formula, is of particular importance because it refers to
the rules according to which the conversion of votes into assembly seats is
conducted. The literature on the effects exerted by electoral systems upon
democratic party systems is voluminous, which suggests that scholarship on
comparative authoritarian institutionalism cannot ignore this substantive
focus. In fact, however, research on electoral system effects under authori-
tarianism is in its infancy.
The theoretical approach of this chapter is the same as in the previous
one, even though the well-known characteristic of electoral rules as the most
easily manipulated component of political systems (Sartori, 1986) suggests
that this time, I have to deal with particularly endogenous institutions. The
empirical analysis replicates the models used in the previous chapters. The
chapter starts with a brief discussion of the theoretically predictable effects
of electoral systems on party system properties. The second section identifies
169
b4407_Ch07.indd 169 1/6/2022 9:16:02 AM

6"×9"
the varieties of the electoral rules used in conjunction with the party systems
examined in this study and formulates a number of hypotheses. The final
section reports empirical findings.
Electoral Systems and Their Effects

Electoral systems are diverse, but there are two main categories reflecting
the basic principles that underlie them, majoritarian and proportional,
and an additional category that is referred to, in a self-explanatory way, as
mixed electoral systems. Each of the three main categories includes several
subcategories that have to be briefly addressed here. Prior to drawing such
distinctions, however, it is important to illuminate the difference between
the two main varieties of electoral rules.
The mechanisms of vote-to-seat conversion lie at the core of contem-
porary research on electoral systems and their political effects. The prop-
osition or “law” of Duverger (1954, p. 127) applies to the most widespread
variety of majoritarian systems, known as plurality rules in single-member
districts. The Duverger law, stating that these rules favor two-party compe-
tition, is often considered as one of the theoretical foundations of political
science (Riker, 1982). According to Duverger, single-member plurality
(first-past-the-post) rules give a representational bonus to large parties and
severely reduce the seat shares of small parties. This aspect is referred to as
a mechanical effect. In this way, plurality rules create a situation in which
only two parties are viable, thus exerting an impact, dubbed psychological
effect, upon the behavior of voters and political elites, as a result of which
the trend toward bipartisan competition consolidates. In the subsequent
literature, starting with the work of Rae (1967), the law of Duverger has
been subjected to rigorous testing, showing that the law tends to hold in
democracies (Powell, 1982; Blais and Carty, 1987; Taagepera and Shugart,
1989; Lijphart, 1994; Benoit, 2001; Clark and Golder, 2006; Golosov, 2017).
At the same time, many scholars noticed that single-member plurality
rules did not necessarily produce the Duvergerian effects, and one of the most
b4407_Ch07.indd 170 1/6/2022 9:16:02 AM

Authoritarian Party Systems and Electoral Systems 171
common explanations of these exceptional cases points to the small parties’

ability to preserve their territorial bases of support as a key to their survival
(Sartori, 1986; Farrell, 2001; Norris, 2004; Meguid, 2008). This factor has
been taken into account in a number of attempts to specify the conditions
under which Duverger’s law is valid formally (Riker, 1982; Kim and Ohn,
1992; Cox, 1997) and to specify them empirically (Golosov, 2018). Thus, in
its expanded version, the Duverger law is relevant not only to party system
fragmentation/format but also to party system nationalization. By implying
that single-member plurality rules reduce the incentives for the emergence
of new parties, the law is also relevant to party system volatility (Ordeshook
and Shvetsova, 1994; Kuenzi and Lambright, 2001; Tavits, 2006), even though
the available empirical evidence in support of this theory is insufficient.
The original proposition of Duverger (1954) was formulated in con-
junction with his theoretical expectation that proportional representation
leads to a rise in the number of parties. Since nearly all empirical studies of
electoral system effects employed district magnitude as the main instrument
for drawing the line between majoritarian and proportional electoral systems,
it can be assumed that the results of these studies — as long as they confirm
the Duverger law in its basic formulation — concur with this expectation
as well. Note that while Duverger formulated his law with special reference
to two-party competition, the empirical evidence in support of the law and
the related propositions has been obtained in a way that does not make a
distinction between bipartism and predominant party systems. In fact, what
has been established empirically in most of the literature is that plurality
systems are linked not to bipartisan competition per se, but rather to low
party system fragmentation. However, this does not pose a major problem
in research on democracies because of the relative rarity of predominant
party systems.
Of course, the world of electoral systems is diverse, and its diversity
cannot be reduced to a simple dichotomy of plurality and proportional rep-
resentation systems. With reference to authoritarianism, making distinctions
within these broad categories is all the more important because, as it will be
b4407_Ch07.indd 171 1/6/2022 9:16:02 AM

6"×9"
shown in what follows, authoritarian regimes often opt for rare and unusual
varieties of electoral rules. In the following overview, I will address only
those varieties that have been used in conjunction with the authoritarian
party systems included in this study.
The category of majoritarian electoral systems comprises, in addition
to single-member plurality, several rather widespread subcategories. One
of them is the two-round majority system, also known as runoff voting, in
which elections can be won only on the condition that the leading candidate
or party receives more than a half of the votes. If none of them reaches this
target, then a second round of voting takes place, normally with the partic-
ipation of most successful candidates from the first round. Duverger (1954)
reasoned that two-round majority systems are favorable for multipartism.
However, as noted by Taagepera and Grofman (1985, p. 342), “because of the
use of single-seat districts, majority runoff may behave more like a plurality
system than a PR [proportional representation] system,” and indeed, this
particular expectation of Duverger has not been supported by cross-national
empirical evidence.
Majoritarian systems are not necessarily used in single-member dis-
tricts, though. In fact, at the dawn of democracy, multimember majoritarian
systems were quite widespread (Colomer, 2007). The most common form of
these rules is multimember plurality system, which is the term used in this
study. It is also referred to as the block vote, plurality at-large, and multiple
non-transferable vote. Elections in such systems are held in districts of greater
than one magnitude in which each voter is entitled to cast as many votes as
there are seats to be filled. One of the properties of multimember plurality
system is that to a greater extent than single-member plurality, it is likely to
result in huge overrepresentation of the leading party. The common term
applied to these shortcomings is the “sweep effect.”
The sweep effect manifests itself in that the strongest party in a majority
of districts sees its full slate of candidates elected (Farrell, 2001, pp. 45–46),
which results in a landslide in terms of the resulting seat allocation. This effect
has been explained theoretically (Taagepera and Shugart, 1989; Grofman,
b4407_Ch07.indd 172 1/6/2022 9:16:02 AM

2006) and observed in several national contexts (Zagarri, 1987; Scarrow,

1999; Benoit, 2001; Calabrese, 2007). An even higher level of overrepresen-
tation can be delivered to the leading party by the party block vote system
(Reilly, 2007). Under this system, each of the electors in multimember
districts is entitled to cast only one vote for a party list, rather than for an
individual candidate, much in the same way as it happens under the most
common form of party-list proportional representation system. But the seats
are not allocated proportionally. Instead, the leading list takes all seats in
those districts where it wins a plurality of votes or, in some instances, an
absolute majority of votes. It has to be mentioned that the sweep effect is
absent if votes in multimember districts are cast for individual candidates
rather than party lists, and the number of votes cast by each of the electors is
smaller than district magnitude. Such systems are referred to as limited vote
systems. In the contemporary world, they survive exclusively in the form of
the so-called single non-transferable vote. This system, under which only one
vote can be cast by each of the voters, tends to yield relatively proportional
seat allocations (Hsieh and Niemi, 1999).
It is common to characterize the methods of proportional representation
with reference to different mathematical algorithms used for seat allocation
(Gallagher, 1992; Benoit, 2000). These formulas can be consequential because
they yield varying degrees of proportionality in vote-to-seat conversion.
However, the effects the mathematical algorithms are conditioned by district
magnitude. They become visible only if district magnitude is relatively small,
which concurs with the common understanding of district magnitude as
the crucial parameter of electoral systems (Taagepera and Shugart, 1989).
Other important differences among proportional representation systems
are related to ballot structure (Teorell and Lindstedt, 2010). However, such
differences are nearly absent among the electoral systems used in countries
with authoritarian party systems. Preferential voting in the form of rank
ordering of party lists has not been employed in any of them, while an open-
list system, being quite consequential in authoritarian Brazil (Mainwaring,
1991), has not had much currency in other non-democratic contexts. As a
b4407_Ch07.indd 173 1/6/2022 9:16:02 AM

6"×9"
rule, proportional representation is implemented in its simplest and most

widespread form known as the closed party-list system.
The definition of mixed electoral systems used in this study is the one
that has been proposed by Massicotte and Blais (1999, p. 345): “We consider
an electoral system to be mixed if its mechanics involves the combination of
different electoral formulas (plurality or PR; majority or PR) for an election to
a single body.” This definition does not differ much from those of Shugart and
Wattenberg (2001) and Reynolds et al. (2005). Mixed electoral systems are
very diverse, and they are particularly diverse in non-democratic countries.
It is notable, however, that none of the authoritarian party systems included
in this study has used a mixed-member proportional system in which the
proportional tier compensates for the low level of proportionality in the
majoritarian tier. This system produces fairly proportional results (Shugart
and Wattenberg, 2001).
In one variety of mixed electoral systems, majoritarian rules are used
in the electoral districts covering a part of the territory, while proportional
representation is used in the other set of territorially defined districts. Such
rules are often referred to as mixed coexistence. The second widespread
variety of mixed rules is known as mixed-superposition (Massicotte and
Blais, 1999), parallel (Reynolds et al., 2005), or mixed-member-majoritarian
(Thames and Edwards, 2006) electoral system. Under this arrangement, all
electors are subjected to the two formulas, which involves creating not only
independent sets of districts for each of the formulas but also endowing
each of the electors with two votes. In some cases, however, proportional
allocation of seats is performed on the basis of the votes cast in the majori-
tarian tire of the system. There are instances when opposition parties receive
guaranteed shares of seats irrespective of the election results, and these seats
are distributed among them on a proportional basis. Finally, in some of
the systems based on the party block vote rules, proportional allocation of
seats can be performed at the district level on the condition that none of the
parties has received a majority of the votes in the given district; otherwise,
a majoritarian principle is applied.
b4407_Ch07.indd 174 1/6/2022 9:16:02 AM

Electoral Systems and the Properties of

Table 7.1 contains information about the electoral systems used in the
countries with authoritarian party systems. The table does not report district
magnitudes. Instead, it uses a trichotomy of small, medium-size, and large
multimember districts as a proxy. In line with Carey and Hix (2011, p. 393)
who identify similarly constructed but more numerous categories on the
basis of substantive considerations, the categories are defined by district
magnitudes of two to five, six to ten, and greater than ten, respectively. Exact
district magnitudes are reported if they are constant across all districts of
the given type.
It has to be noted that the reason why district magnitude is not employed
as an explanatory variable in the following analysis is not so much the
unavailability of the relevant data, which could have been remedied by using
numerical proxy values, but rather the widespread use of multimember
majoritarian systems, sometimes in the districts of fairly large magnitudes,
under authoritarianism. There is a widespread theory that in such systems
greater district magnitude decreases proportionality, thus affecting party
system parameters (Blais and Carty, 1987; Taagepera and Shugart, 1989;
Grofman, 2006). While the empirical evidence in support of this theory is
not entirely consistent (Benoit, 2001), and there are studies specifying con-
ditions under which the theory may not hold (Golosov, 2003a; Eggers and
Fouirnaies, 2014), there are reasons to expect that district magnitude is not
a reliable tool in the statistical analyses of authoritarian electoral politics.
The table shows that authoritarian electoral systems are strikingly
diverse, but there are two regularities that differentiate the universes of
authoritarian and democratic electoral systems. According to Bormann
and Golder (2013, p. 363), proportional, majoritarian, and mixed electoral
systems were employed in about 50 percent, 42 percent, and 8 percent
of democratic elections, respectively, in the 1950s, and in 49 percent,
33 percent, and 18 percent, respectively, by the 2000s. The data at my
b4407_Ch07.indd 175 1/6/2022 9:16:02 AM

6"×9"
Table 7.1. Electoral systems used in the countries with authoritarian

party systems.
Years of elections (if changed during the period)/
Party system electoral system description
Algeria Proportional representation in multimember districts of
small to large magnitude
Angola Proportional representation in five-member multimember
districts and a large nationwide multimember district
Armenia 2003–2012, mixed: plurality in single-member districts
and proportional representation in a large nationwide
multimember district; 2017, proportional representation
in a large nationwide multimember district
Azerbaijan 1995–2000, mixed: plurality in single-member districts and
proportional representation in a large nation-multimember
district; 2005–2015, plurality in single-member districts
Bolivia Proportional representation in large multimember districts
Brazil Proportional representation in multimember districts of
small to large magnitude
Burundi Proportional representation in large multimember districts
Cambodia Proportional representation in multimember districts of
small to medium magnitude
Cameroon Mixed: plurality in single-member districts, party block vote
on the basis of two-round majority in small to medium-size
multimember districts, with possibility of proportional seat
allocation conditional upon election results
Chad Mixed: two-round majority in single-member districts and
party block vote in small to medium-size multimember
districts, with possibility of proportional seat allocation
conditional upon election results
Colombia 1945–1949, proportional representation in small to large
multimember districts; 1951, mixed: party block vote in
small to large multimember districts, with possibility of
proportional seat allocation conditional upon election results
Republic of Congo Two-round majority is single-member districts
Cote d’Ivoire Plurality in single-member districts and party block vote in
small to medium-size multimember districts
Djibouti 1992–2008, party block vote in small to large multimember
districts; 2013–2018, mixed: party block vote in small to
large multimember districts, with possibility of proportional
seat allocation conditional upon election results
b4407_Ch07.indd 176 1/6/2022 9:16:02 AM

Dominican Republic Proportional representation in small to medium-size
multimember districts
Egypt 1976–1979 and 1987–2010, plurality in small to
medium-size multimember districts; 1984, proportional
representation in small to large multimember districts
El Salvador 1961, party block vote in small to large multimember
(1961–1978) districts; 1964–1978, proportional representation in small
to large multimember districts
El Salvador Proportional representation in small to large multimember
(1982–1991) districts
Equatorial Guinea Proportional representation in small to medium-size
Ethiopia Plurality in single-member districts
Gabon Two-round majority in single-member districts
Gambia Plurality in single-member districts
Guatemala Proportional representation in small to medium-size
Guyana Proportional representation in a large nationwide
multimember district
Indonesia Proportional representation in small to large multimember
districts
Iran Plurality in single-member districts and small multimember
districts
Kazakhstan 1999–2004, mixed: two-round majority in single-member
districts and proportional representation in a medium-size
nationwide multimember district; 2007–2016, proportional
representation in a large nationwide multimember district
Madagascar Mixed: party block vote in small to large multimember
districts, with possibility of proportional seat allocation
conditional upon election results
Malaysia Plurality in single-member districts
Mauritania Mixed: two-round majority in single-member districts and
party block vote in two-member multimember districts, with
possibility of proportional seat allocation conditional upon
election results
(Continued)
b4407_Ch07.indd 177 1/6/2022 9:16:02 AM

6"×9"
Mexico 1946–1976, plurality in single-member districts; 1979–1988,
mixed: plurality in single-member districts and proportional
representation in large multimember districts
Morocco 1984–1993, plurality in single-member districts;
1997–2016, proportional representation in small to
medium-size multimember districts and, starting from 2011,
in a large nationwide multimember district
Nepal Plurality in single-member districts
Nicaragua Mixed: party block vote in small to large multimember
districts with a provision for minority guarantee seats
allocated on a proportional basis
Paraguay Mixed: party block vote in small to large multimember
districts with a provision for minority guarantee seats
allocated on a proportional basis
Philippines Plurality in single-member districts and small multimember
districts
Portugal Party block vote in small to medium-size multimember
districts
Russia 2003 and 2016, mixed: plurality in single-member
districts and proportional representation in a large
nationwide multimember district; 2007–2011, proportional
Rwanda Proportional representation in a large nationwide
Serbia Proportional representation in a large nationwide
Seychelles Mixed: plurality in single-member districts with additional
seats allocated by proportional representation on the basis
of single-member district vote
Sierra Leone Plurality in single-member districts
Singapore Plurality in single-member districts and, starting from 1984,
also in small to medium-size multimember districts
South Africa Plurality in single-member districts
South Korea 1963–1971, mixed: plurality in single-member districts with
additional seats allocated by proportional representation
on the basis of single-member district vote with a majority
bonus for the leading party; 1973–1978, single non-
transferrable vote in two-member districts
b4407_Ch07.indd 178 1/6/2022 9:16:02 AM

South Vietnam Plurality in single-member districts
Sudan 2000, plurality in single-member districts; 2010–2015,
mixed: plurality in single-member districts and proportional
representation in small to medium-size multimember
districts
Taiwan 1972–1989, single non-transferrable vote in small to
medium-size multimember districts; 1992, mixed: single
non-transferrable vote in small to medium-size multimember
districts with additional seats allocated by proportional
representation on the basis of multimember district vote
Tajikistan Mixed: plurality in single-member districts and proportional
Thailand Plurality in small to medium-size multimember districts
Togo 1994–2002, two-round majority in single-member districts;
2007, proportional representation in small multimember
districts
Tunisia 1981–1989, party block vote in medium-size multimember
districts; 1994–2009, mixed: party block vote in medium-
size multimember districts with a provision for minority
guarantee seats allocated on a proportional basis
Uganda Plurality in single-member districts
Uzbekistan Two-round majority in single-member districts
Yemen Plurality in single-member districts
Zimbabwe Plurality in single-member districts
(Rhodesia)
Zimbabwe 1985–2008, plurality in single-member districts; 2013–2018,
mixed: plurality in single-member districts with additional
seats allocated by proportional representation on the basis
of single-member district vote
disposal are cross-temporal, covering the whole period from 1945–2018,

and the observations are system-level rather than election-level. But the
observed difference is significant enough to be taken into account: among
the 74 electoral systems in the table, 28 percent are proportional, 42 percent
are majoritarian, and 30 percent are mixed. Thus, autocracies have a strong
b4407_Ch07.indd 179 1/6/2022 9:16:02 AM

6"×9"
preference toward majoritarian systems, and a very well-articulated prefer-

ence toward mixed systems.
The second difference is related to the relative spread of multimember
majoritarian systems. In democracies, such systems are the things of the past,
often properly viewed as a part of the predemocratic electoral institutions of
19th century Europe (Ahmed, 2010). Even though multimember majoritar-
ian systems are not rare at the sub-national level, where they survive mostly
for technical reasons (Van der Kolk, 2007; Golosov and Turchenko, 2021),
they are almost extinct at the national level of democratic politics. Recently,
such a system has been used to fill the whole lower chamber of the national
legislature in only one long-standing democracy, Mauritius (Kasenally,
2011), and a variety of systems of this type are employed in several small
island jurisdictions where party systems are absent (Levine and Roberts,
2005). Under authoritarianism, the number of such systems has declined
over time, but it remains substantial.
Of course, some of the peculiarities of authoritarian electoral systems
can be related to territorial diffusion. The use of complex mixed systems
with multimember plurality components in some of the African countries
can be traced to the attempts of electoral engineering in the French Fourth
Republic (Browne and Hamm, 1996), while the spread of mixed systems in
the post-Soviet countries was influenced by the adoption of such a system
in Russia (Lundberg, 2009). It is also clear that some of the authoritarian
regimes simply inherited their electoral systems from the previous political
conditions and, given that these systems did not prevent them from con-
solidating the authoritarian rule, had little reason to abandon them. At the
same time, the observed difference between democratic and authoritarian
electoral systems supports the argument that purposeful electoral engineer-
ing is not alien to autocracies. In the recent literature, the non-systematic
evidence on the authoritarian regimes’ willingness and ability to use this tool
is quite abundant (Lee, 1999; Pripstein Posusney, 2002; Golosov, 2006a), and
some cross-national and/or quantitative studies point in the same direction
(Golosov, 2016d; Turchenko, 2020).
b4407_Ch07.indd 180 1/6/2022 9:16:02 AM

The reason for the preference for majoritarian systems displayed by

autocracies can be rather confidently related to the main political con-
sequence of such systems, a reduction in the effective number of parties
stemming from the mechanical and psychological Duvergerian effects. The
“sweep effect” of the multimember majoritarian rules reduces party system
fragmentation to an even greater extent. Indeed, a key element within the
incentive structure of authoritarian elections is the autocrat’s desire to win
by a large margin, and this can be delivered by majoritarian electoral systems
with a much greater degree of certainty than by proportional representation.
The negative association between majoritarian electoral systems and
two other party system parameters, volatility and nationalization, has been
theoretically explained above. The presence of territorially based parties is
not in contradiction with the incentive structure of authoritarian elections
because such parties are normally small, thus posing no threat to the political
monopoly at the national level, and they can be useful for power sharing
between the national and regional elites. Of course, the sustained dominance
of the leading pro-regime party presupposes rather high party system sta-
bility. At the same time, the permitted opposition parties are expendable
for the regime and thus can be swept away as a result of volatility. Overall,
these considerations make it possible to expect that the effects of majoritar-
ian electoral systems on party system parameters under authoritarianism
will be the same in direction, if not in scope, to what has been observed in
democracies. Hence, I hypothesize that:
H1. Majoritarian electoral systems will be associated with party

systems characterized by low fragmentation, low volatility, and
low nationalization.
As noted above, proportional representation is not very popular among

authoritarian regimes. However, it has been used in many autocracies, which
can be partly explained with reference to the institutional legacies of the pre-
vious regimes and diffusion. At the same time, party system fragmentation is
b4407_Ch07.indd 181 1/6/2022 9:16:02 AM

6"×9"
not problematic for some of the authoritarian regimes. The peculiar incentive
structures of such regime genomes and institutional arrangements have been
discussed elsewhere in this book. Such incentive structures are undoubtedly
consistent with the use of proportional representation.
Irrespective of the peculiarities related to the regime genome or patterns
of executive–legislative relations, the incentive structure of authoritarian
elections does include securing some level of representation for the permit-
ted opposition, which can be achieved by proportional allocation of seats
in the conditions when the likelihood of electing a single-party assembly
by majoritarian rules is high. Several overtly authoritarian regimes, such
as Nicaragua, Paraguay, and Tunisia, solved this problem by establishing
legal provisions for minority guarantee seats allocated on a proportional
basis. These provisions were built into mixed electoral systems, though.
The experience of several other countries suggests that the incentives for
using proportional representation under authoritarianism can be different.
Perhaps the most extreme example is Kazakhstan where the first elections
held by proportional representation rules created a legislature with all seats
filled exclusively by the dominant pro-regime party, the Radiant Fatherland,
which clearly suggests that the electoral reform did not seek more comfort-
able conditions for the permitted opposition parties.
The case of Russia is similarly illustrative. In 1993–2004, the country
used a mixed parallel electoral system. Under these rules, the country’s party
system was fragmented, which was of course possible to change by shifting
to purely majoritarian rules. However, such a shift could have aggravated
a different problem faced by the national authorities. The mixed system
opened some space for the advocacy of regional interests, which was one
of the primary goals pursued by independent deputies elected in single-
member districts (Shevchenko and Golosov, 2001). Personal political
resources accumulated by such deputies were substantial. Neither the depu-
ties themselves nor their regional patrons were eager to put these resources to
the service of the national authorities (Golosov, 2002). After the emergence
of the dominant pro-regime party, United Russia, this problem was resolved,
b4407_Ch07.indd 182 1/6/2022 9:16:02 AM

quite successfully, by creating institutional environments that excluded

independent sub-national political activism. In 2007, the introduction of
proportional representation simply eliminated independent candidacy as
a route to electoral success. At the same time, the electoral reform greatly
increased the role of the central governing bodies of United Russia, and
thereby of the national executive, in determining the personal composition
of the legislature. Parallel developments occurred at the sub-national level
where elements of proportional representation were built into all systems of
regional legislative elections (Golosov, 2013b). In this way, the organizational
cohesion of the dominant party increased, which is a commonly observed
effect of proportional representation (Cox, 1990).
Of course, the cohesion of the dominant party is a matter of major
concern for any electoral authoritarian regime. Thus, the gains achieved by
the shift to proportional representation were related not to the structure of
the party system but rather to the organizational development of the dom-
inant party. The risk of increasing party system fragmentation was tempo-
rarily neutralized by introducing restrictive legislation on political parties,
as a result of which many of them ceased to exist, and by setting the legal
threshold of representation at the unusually high level of 7 percent. At the
regional level, where the magnitudes of proportional representation districts
were relatively low, this was supplemented by the wide use of a particularly
disproportional mode of seat allocation, the Imperiali highest averages
method (Golosov, 2014a). It is however important to mention that at the
national level, Russia continued to use a highly proportional seat allocation
formula, the Hare quota and largest remainder method, in a nationwide
district of huge magnitude, 450.
In 2011, after facing less than desirable results in the national legisla-
tive elections that brought Russia’s party system to a relatively high level
of fragmentation, the Russian authorities announced that they would
return to the mixed system, which was implemented in the 2016 national
legislative elections. By that time, however, the national executive had
already successfully solved the problems that were sought to be dealt with
b4407_Ch07.indd 183 1/6/2022 9:16:02 AM

6"×9"
by introducing a pure proportional representation system: resourceful

independent candidates almost entirely disappeared, and a vast majority
of seats in the majoritarian section of the 2016 elections were filled by the
nominees of United Russia (Kynev, 2017). Note that fragmentation in the
proportional section of the elections remained high enough, but within
the context of mixed electoral rules, this natural effect of proportional
representation became politically inconsequential. As argued in one study,
the increased fragmentation was even beneficial for the authorities in terms
of enhancing their capacities for co-opting the opposition and creating a
more favorable image of “truly competitive” elections (Wilson, 2016). With
regards to the use of proportional representation in very different political
environments of Burundi and Rwanda, Stroh (2010, p. 1) argues that pro-
portional representation “impedes the local accountability of politicians
in a political environment in which the government is not held in check
by a democratic opposition.”
The discussion above leads me to a preliminary conclusion that to a
greater extent than those of all other institutional arrangements, the effects
of authoritarian electoral systems can be linked to the endogenous character
of these institutions. Proportional representation facilitates party system
fragmentation. As a result, it tends to be employed not only in the condi-
tions when the threats of multipartism to regime stability are considered
as negligibly small, but also when the organizational consolidation of the
dominant party stands high on the regime’s agenda.
At the same time, the benefits of proportional representation for
autocracy can be considerable, be they in the form of projecting a more
convincing image of multiparty competition or, more likely, in the form of
enhancing the organizational cohesion and national appeal of pro-regime
parties. Obviously, some of these benefits are associated with high party
system nationalization. In theory, proportional representation allows for
greater levels of party system volatility than is the case under majoritarian
systems. As long as volatility manifests itself primarily in the high turnover
of opposition parties, so that none of them is able to consolidate its grip
b4407_Ch07.indd 184 1/6/2022 9:16:02 AM

on the electorate and achieve organizational stabilization, this effect can be

beneficial for autocracy. Hence, I hypothesize that:
H2. Proportional representation systems will be associated with party

systems characterized by high fragmentation, high volatility, and
high nationalization.
As demonstrated by an example above, the autocrats may engage in

electoral engineering by changing the rules of the game at will. Similar
benefits can be however obtained by creating complex electoral systems in
which the potentially dangerous effects of their components can be neutral-
ized, while their benefits to the autocrats can be maximized. This applies to
mixed electoral systems. Sometimes praised as taking the best from both
worlds of electoral systems (Shugart and Wattenberg, 2001), which might
be true or not with regards to democracies (for different views, see Linhart
et al., 2019; Jastramskis, 2019), such systems, as demonstrated above, are
particularly widely employed by authoritarian regimes. Indeed, as noted by
Blais and Massicotte (2002, p. 54), it was for a long time common to dismiss
such systems as “instances of sheer manipulation.”
Of course, the diversity of mixed electoral systems does not allow for
a theoretical prediction that their effects will be uniform. In some of them,
the elements of proportional representation are too small to make a differ-
ence. In Seychelles and Zimbabwe, the proportional allocation of relatively
small numbers of additional seats is performed on the basis of votes cast
in majoritarian districts, which obviously reduces the effects of propor-
tional representation. The complex electoral rules of Cameroon, Chad, and
Mauritania effectively restrict the use of proportional seat allocation to a
very small number of districts. In Djibouti and Tunisia, the proportional
allocation of additional seats has been performed at the national level, but
the numbers of such seats are very small. There is no reason to expect that
in such conditions, the effects of mixed electoral systems will be significantly
different from those of purely majoritarian rules, even though each of the
b4407_Ch07.indd 185 1/6/2022 9:16:02 AM

6"×9"
above-mentioned modifications slightly increases the potential scope of

representation of opposition parties.
Mixed-superposition electoral systems stand out in that the numbers
of seats allocated on the proportional basis are normally large enough, and
these seats tend to be allocated in large-magnitude districts. However, even
in this case it would be logical to expect that the effects of such systems
will be intermediate between those of the pure forms of proportional and
majoritarian rules. Given that these effects are opposite to each other, this
equals to the prediction that the statistical analysis will be able to establish
no discernible effects of mixed electoral systems whatsoever.
There is a stream of research according to which even in the absence
of institutional linkage between the majoritarian and proportional sections
of these systems, the effective linkages may occur as a result of the strategic
choices made by the involved political actors or individual voters. In par-
ticular, such linkages, referred to as “contamination effects” or “interaction
effects” by those scholars who discovered them (Herron and Nishikawa, 2001;
Cox and Schoppa, 2002; Ferrara et al., 2005), may lead to the proliferation
of parties in the proportional section of elections. Even in democracies,
however, the empirical evidence supporting the existence of this effect is
scarce. The literature identifies several conditions under which they are
likely to occur (Riera, 2013; Ragauskas and Thames, 2020), but such con-
ditions are normally absent in autocracies. The above considerations make
me hypothesize that:
H3. The effects of mixed electoral systems will be similar to those of

the majoritarian systems.
Empirical Findings
Table 7.2 reports the descriptive information about the parameters of the
authoritarian party systems in relation to electoral systems, while Tables 7.3
and 7.4 present the results of the statistical analysis with regards to party
b4407_Ch07.indd 186 1/6/2022 9:16:02 AM

Table 7.2. Party system parameters in relation to electoral systems.

Standard
Fragmentation, majoritarian 1.64 1.22 1.00 8.75 1.12 150
systems
Fragmentation, proportional 2.14 1.55 1.00 11.80 1.87 84
representation systems
Fragmentation, mixed 1.56 1.48 1.00 4.35 0.53 60
systems
Volatility, majoritarian 13.02 8.77 0.00 58.21 13.91 121
systems
Volatility, proportional 18.54 16.42 0.51 64.99 13.41 69
Volatility, mixed systems 15.93 13.45 0.00 52.27 13.26 47
Nationalization, majoritarian 0.64 0.70 0.29 0.94 0.19 19
systems
Nationalization, proportional 0.80 0.84 0.46 0.98 0.15 16
Nationalization, mixed 0.70 0.66 0.48 0.89 0.15 7
systems
Table 7.3. Summary of ANOVA comparison of countries with different

electoral systems by fragmentation.
Between groups 16.44 2 8.22 4.85 0.01
Within groups 493.38 291 1.70 — —
Total 509.83 293 — — —
system fragmentation. The data in the tables clearly support the above-
formulated hypotheses by demonstrating that the levels of party system
fragmentation in the presence of proportional representation are signifi-
cantly higher than under majoritarian and mixed rules. Note that mixed
electoral systems are associated with the levels of fragmentation that are
even lower than under purely majoritarian rules, even though the differ-
ence is not significant. At the same time, as follows from Tables 7.5 and 7.6,
b4407_Ch07.indd 187 1/6/2022 9:16:02 AM

6"×9"
Table 7.4. Bonferroni corrected comparisons of countries with different

electoral systems by fragmentation.
Mean Standard
Majoritarian Proportional −0.50 0.18 0.02
representation
Mixed 0.08 0.20 1.00
Proportional Majoritarian 0.50 0.18 0.02
representation Mixed 0.57 0.22 0.03
Mixed Majoritarian −0.08 0.20 1.00
Proportional −0.57 0.22 0.03
representation
Table 7.5. Linear regression results, fragmentation by electoral

systems (n = 294 adjusted for 57 clusters; unstandardized coefficients;
Proportional
Majoritarian representation Mixed
Electoral systems −0.24 (0.24) 0.53 (0.38) −0.30 (0.24)
Index of electoral democracy 2.11* (1.22) 2.34* (1.19) 2.36* (1.30)
Index of ethnic fractionalization −0.07 (0.27) −0.06 (0.27) −0.17 (0.29)
Second elections −0.20** (0.10) −0.21** (0.10) −0.19* (0.10)
Real GDP per capita in 2011 −0.02** (0.01) −0.02** (0.01) −0.02** (0.01)
USD equivalent (thousands)
Constant 1.47*** (0.40) 1.17*** (0.29) 1.41*** (0.36)
R-squared 0.03 0.06 0.03
at 0.1 level.
electoral systems are not reliable predictors of party system fragmentation

under authoritarianism. Even if the differences among their effects are
visible enough, these differences do not convert into statistically significant
cross-system variations on the parameter of party system fragmentation.
Figure 7.1 and Table 7.7 present the results of the analysis with regards
to party system format. Consistent with my expectations, proportional
b4407_Ch07.indd 188 1/6/2022 9:16:02 AM

Table 7.6. Multilevel mixed-effects regression results, party system

fragmentation by electoral systems.
Standard
Majoritarian −0.17 0.17 0.32
Proportional representation 0.10 0.21 0.64
Index of ethnic fractionalization −0.19 0.58 0.75
Constant 1.62 0.43 0.00
Country 1.06 0.10 —
Year 0.00 0.00 —
Note: Wald chi2 = 8.87; Log restricted-likelihood = −367.13; Prob > chi2 = 0.26.
representation systems are strongly associated with multipartism. But,

rather unexpectedly, Table 7.7 demonstrates that proportional repre-
sentation is strongly but negatively associated with predominant party
systems with fragmented opposition, while the associations between this
party format and two other varieties of electoral rules are insignificant but
positive. Judging from the theoretically predictable effects of proportional
representation, as they have been identified in the main body of research on
democracies, this should not be the case. Note, however, that when intro-
ducing this electoral system, the autocrats take severe precautions against
the possible proliferation of political parties — as is evident from the case
of Russia and can be supported by evidence from Burundi, Kazakhstan,
and Rwanda as well. Besides, party-list proportional representation in
nationwide electoral districts practically excludes independent candidacy
as a route to electoral success, which also decreases fragmentation within
the opposition camp.
b4407_Ch07.indd 189 1/6/2022 9:16:02 AM

6"×9"
1 1
Plate 1 Plate 2
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1
0.9
Plate 3
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

authoritarian elections by electoral systems. Plate 1: Majoritarian sys-
tems. Plate 2: Proportional representation systems. Plate 3: Mixed
systems.
Tables 7.8–7.11 analyze the impact of electoral rules upon the volatility

of authoritarian party systems. Here again, in accordance with my hypothe-
ses, proportional representation stands out as a system associated with high
levels of volatility, while the effect of majoritarian systems is opposite. Mixed
electoral systems are associated with the levels of volatility that are lower
than those of majoritarian systems, even though the difference is insignif-
icant. Table 7.10 shows that in contrast to what has been observed in the
above analysis of party system fragmentation, majoritarian and proportional
electoral systems do serve as significant predictors of party system volatility
in the theoretically anticipated directions. Table 7.11 clarifies this finding by
b4407_Ch07.indd 190 1/6/2022 9:16:03 AM

Table 7.7. Two-way analysis of variance: electoral systems and party

system formats.
Predominant
party systems Other
with fragmented systems, Pearson Fisher’s
opposition, n n chi2 Cramér’s V exact
Majoritarian 62 88 1.67 0.08 0.23
Proportional 22 62 6.69*** −0.15 0.01
representation
Mixed 22 24 2.35 0.09 0.14
Predominant
party systems Other
with concentrated systems,
opposition, n n
Majoritarian 56 94 0.08 −0.02 0.81
Proportional 32 52 0.00 0.00 1.00
representation
Mixed 16 30 0.25 −0.03 0.74
Other
Systems without systems,
majority parties, n n
Majoritarian 32 118 1.33 −0.07 0.28
Proportional 30 54 8.59*** 0.17 0.00
representation
Mixed 8 38 1.36 −0.07 0.27
at 0.1 level.
Table 7.8. Summary of ANOVA comparison of countries with different

electoral systems by volatility.
Between groups 0.14 2 0.07 3.68 0.03
Within groups 4.36 234 0.02 — —
Total 4.49 236 — — —
b4407_Ch07.indd 191 1/6/2022 9:16:03 AM

6"×9"
Table 7.9. Bonferroni corrected comparisons of countries with different

electoral systems by volatility.
Mean Standard
Majoritarian Proportional representation −0.06 0.02 0.02
Mixed −0.03 0.02 0.65
Proportional Majoritarian 0.06 0.02 0.02
representation Mixed 0.03 0.03 0.94
Mixed Majoritarian 0.03 0.02 0.65
Proportional representation −0.03 0.03 0.94
Table 7.10. Linear regression results, volatility by electoral systems

(n = 294 adjusted for 57 clusters; unstandardized coefficients; clustered-
robust errors in parentheses).
Proportional
Majoritarian representation Mixed
Electoral systems −0.04* (0.02) 0.06** (0.03) −0.004 (0.022)
Index of electoral 0.37** (0.16) 0.40*** (0.15) 0.39** (0.18)
democracy
Index of ethnic 0.03 (0.04) 0.03 (0.04) 0.02 (0.04)
fractionalization
Second elections 0.06*** (0.02) 0.06*** (0.02) 0.06*** (0.02)
Real GDP per capita in −0.002** (0.001) −0.003*** (0.001) −0.002* (0.001)
2011 USD equivalent
(thousands)
Constant 0.07 (0.05) 0.03 (0.04) 0.06 (0.05)
R-squared 0.13 0.14 0.11
at 0.1 level.
demonstrating that if both systems are entered into the regression analysis,
and in the presence of control variables, only majoritarian systems survive
as a significant predictor, and the effect becomes only marginally significant.
Table 7.12 shows pairwise correlations between the dichotomous
variables pertaining to electoral systems and the index of party system
b4407_Ch07.indd 192 1/6/2022 9:16:03 AM


electoral systems.
Standard
Majoritarian −0.04 0.03 0.09
Proportional representation −0.01 0.03 0.78
Index of ethnic fractionalization 0.03 0.05 0.58
Real GDP per capita in 2011 USD 0.00 0.00 0.40
Constant 0.11 0.05 0.05
Country 0.08 0.01 —
Year 0.07 0.01 —

nationalization by electoral systems, n = 42.
chi2 Significance
Majoritarian −0.35 0.02
Proportional representation 0.40 0.01
Mixed −0.05 0.78
nationalization. Despite the small number of observations and rather

rough operationalization of the concept of electoral systems, the observed
associations are statistically significant, which confirms my theoretical
expectations regarding the strong positive association between proportional
representation and party system nationalization and the opposite effects of
majoritarian rules. The association between mixed electoral systems and
party system nationalization is statistically insignificant but, consistent with
H3, it is the same in direction as for majoritarian systems.
b4407_Ch07.indd 193 1/6/2022 9:16:03 AM

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Chapter 8
The Destinations of
S
o far, this book has dealt with the properties of authoritarian party
systems by treating them as belonging to a closed set of system-level
observations defined by a number of simple quantitative criteria. Of
course, the real universe of authoritarian party systems is ever changing. The
purpose of this chapter is to briefly discuss the dynamic aspects of authori-
tarian party development by focusing, first, on the legacies they leave behind
after they cease to exist, and second, on the prospects for the existing ones
and those yet to come. It has to be immediately recognized that, related to
the quantitative methods employed in this study, its retrospective nature
precludes a definitive assessment of the accuracy of predictions. One of the
reasons is that some of the trends in the development of authoritarian party
systems are still in the making, and the amount of the relevant data is small.
The very feasibility of statistical inference in such conditions is limited. This
problem, however, is inherent in any branch of science that has to rely upon
statistical/probabilistic rather than experimental methods (Peters, 2020), and
it can be partly solved by specifying the conditions under which theoretical
expectations can hold.
The Legacies of Authoritarian Party Systems

A majority of the authoritarian party systems included in this study — 37
out of 57 — are already extinct. The legacies of these systems, in the form of
the impact exerted by them upon the properties of party systems that have
195
b4407_Ch08.indd 195 1/6/2022 9:16:18 AM

6"×9"
come to replace them, can be expected to be lasting, if only because of the

well-described in the literature tendency of the origins of political parties
to exert a decisive impact on their further development (Panebianco, 1988).
There is a rapidly growing stream of scholarly literature on what is called
“authoritarian successor parties,” understood as those ruling parties of the
defunct authoritarian regimes that have accepted democracy and comply
with the norms of free political competition.
This stream of research has generated a number of studies, both broadly
cross-national (Grzymala-Busse, 2019; Morse, 2020) and with more nar-
row regional foci (Hicken and Kuhonta, 2011; Riedl, 2014; Loxton and
Mainwaring, 2018; Kasuya and Sawasdee, 2019). Most of these studies deal
exclusively with authoritarian successor parties. A recent study of Grzymala-
Busse (2020) takes a more systemic perspective, but this study also employs
the political strategies of authoritarian successor parties, in the form of their
exit from power and subsequent “reinvention” as committed democratic
competitors, as main explanatory variables in the analysis of the properties of
post-authoritarian party systems (see also Dinas and Northmore-Ball, 2020).
The findings generated by this stream of literature are important, but
the exclusive focus on authoritarian successor parties and their strategies is
not consistent with the purposes of this study with its main perspective on
party systems rather than on their individual components. Fortunately, a
wider systemic approach can be informed by insights from an earlier body
of research on the so-called communist successor parties in the countries
of East Central Europe and the former Soviet Union. The resurgence of the
former ruling parties in these countries naturally stimulated scholarly interest
in them (for overviews, see Bozóki and Ishiyama, 2002; Grzymala-Busse,
2002). Quite naturally, this literature was similar to the more recent studies
in its focus on the destinations of individual communist parties rather than
party systems. Indeed, no other parties, with the exception of the satellite
“fellow traveler” parties most of which perished after the fall of communism
(Baylis, 1998), were in existence under the communist regimes. But exactly
for this reason, the literature on communist successor parties had to pay
b4407_Ch08.indd 196 1/6/2022 9:16:18 AM

The Destinations of Authoritarian Party Systems 197
closer attention to the macro-political factors that determined the shapes

of the emerging post-communist party systems.
Two such factors are of particular interest for this study: the type of
the previous regime and the mode of regime transition (Kitschelt, 1995;
Golosov, 1996; Rivera, 1996; Ishiyama, 1999). As of today, the historical cat-
egorizations of communist regime types that can be found in this literature
present a narrow interest. From the methodological point of view, however,
this approach can be viewed as an early formulation of what is referred to
in this book as the regime genome, which is one of the central concepts
of this study. In the previous chapters, regime genomes have been found
highly consequential for the parameters of authoritarian party systems,
which implies that they can be consequential in shaping their legacies as
well. This implication forms the main context of some of the hypotheses to
be tested in what follows.
The second factor, the mode of regime transition, has not been discussed
above, and it deserves some additional attention. The concept of regime
genome implies that authoritarian party systems are intrinsically linked to
the political regimes that generate them. Often, albeit not necessarily, these
party systems experience deep transformations after regime change. In this
respect, they are not different from democratic party systems that often
assume different shapes after authoritarian interludes (Valenzuela et al.,
2018). One of the key concepts of “transitology,” the “pacted transition” as
developed in the foundational work of O’Donnell and Schmitter (1986),
referred to the phenomenon that was identified as a crucial determinant not
only in respect to the overall patterns of political development after demo-
cratic transitions, but also, specifically, in respect to party system formation
(Karl, 1990; Karl and Schmitter, 1991). Thus, the mode of regime transition
can be viewed as an important intermediate variable that affects the way in
which authoritarian legacies are incorporated into the new political order.
As such, the concept of “pacted transition” has found little use outside
of research on Southern Europe and Latin America (Bunce, 1995). In this
study, I use a different but broadly related concept of “soft transition.” It is
b4407_Ch08.indd 197 1/6/2022 9:16:18 AM

6"×9"
defined as a regime transition in which the ruling group of the incumbent

authoritarian regime plays a significant role. The category does include
“pacted transitions” in the traditional understanding of the term, implying
that the ruling elite and counter-elites reach a deal regarding the prospec-
tive rules of the game. But it also refers to situations in which the role of
counter-elites was small, and the primary impetus for political change came
from international pressures, which has been described by an observer of
the political transformation in the Seychelles as “the increasing linkage
of aid to democratic change by Western donors, as well as the significance
of the new international climate” (Hatchard, 1993, p. 602).
In fact, some of the earliest Latin American democratizations of the
“third wave” clearly deviated from the trajectory of pacted transition by
being achieved with very little deal-making (Conaghan and Espinal, 1990).
The recent literature suggests that even without international pressures,
authoritarian-led democratizations can be facilitated by factors not related
to the strength of the counter-elite or even inversely related to it because if
the counter-elite is weak, the risks of democratization are not perceived as
existential by the autocrat (Riedl et al., 2020). When building the category
of soft transitions, I do not extend it to those cases when trends toward
liberalization were reversed in a short time, resulting in the reinstallation
or even strengthening of the incumbent authoritarian regimes, and to those
cases when change resulted from the political settlements of civil wars.
Theoretically, such situations can be located at the far ends of the continuum
of elite-driven political change. Their effects on party system development
are expected to be different from those of the soft transitions as such.
The descriptive Table 8.1 provides information about the reasons
for termination of the political regimes conjoined with the extinct party
systems that are included in this study. The overall duration of the party
system is calculated by subtracting the year of the earliest election from the
year of termination or, for the existing systems, from 2019, and adding one.
It has to me mentioned that three of the included party systems, those of
Colombia, Mexico, and Nicaragua, came into existence earlier than 1945,
b4407_Ch08.indd 198 1/6/2022 9:16:18 AM

Table 8.1. Reasons for termination of the political regimes conjoined

with the extinct party systems.
Duration of
the party
Year of system
Country termination (years) Reason for termination
Armenia 2017 15 Electorally induced peaceful revolution
Bolivia 1964 9 Military coup
Brazil 1984 19 Soft transition
Colombia 1953 9 Military coup
Cote d’Ivoire 2002 13 Civil war
Dominican Republic 1978 13 Soft transition
Egypt 2011 36 Popular uprising and military coup
El Salvador (1961–1978) 1979 19 Military coup
El Salvador (1982–1991) 1993 12 Post-civil war settlement
Gabon 2017 28 Short-term liberalization
Gambia 2017 21 Foreign intervention
Guatemala 1982 17 Military coup
Guyana 1992 20 Soft transition
Indonesia 1998 28 Popular uprising and military coup
Iran 1975 20 Transition to a single-party system
Madagascar 1972 13 Military coup
Malaysia 2018 55 Soft transition
Mauritania 2005 14 Military coup
Mexico 1990 45 Soft transition
Nepal 2002 12 Royal coup
Nicaragua 1979 30 Violent revolution
Paraguay 1989 27 Military coup
Philippines 1959 14 Soft transition
Portugal 1974 26 Military coup
Serbia 2000 9 Electorally induced peaceful revolution
Seychelles 2002 10 Soft transition
Sierra Leone 1978 12 Transition to a single-party system
South Africa 1994 47 Post-segregation system settlement
South Korea 1979 17 Military coup
South Vietnam 1963 8 Military coup
Sudan 2019 20 Popular uprising and military coup
(Continued)
b4407_Ch08.indd 199 1/6/2022 9:16:18 AM

6"×9"
Duration of
the party
Year of system
Country termination (years) Reason for termination
Taiwan 1995 24 Soft transition
Thailand 1991 17 Military coup
Togo 2007 14 Short-term liberalization
Tunisia 2011 31 Popular uprising
Yemen 2011 19 Popular uprising
Zimbabwe (Rhodesia) 1979 10 Post-segregation system settlement
so that they were effectively more durable than it is reported in the table.
As is evident from the table, military coups were the most common reason
for the termination of authoritarian party systems (11 cases), followed by
soft transitions (10 cases). Other reasons were diverse, and they are labeled
in self-explanatory ways. Note that in some of the literature, the electorally
induced peaceful revolutions, i.e., those cases when political change occurs
as a result of non-violent protests against perceivably unfair election results,
are referred to as “colored revolutions” (Tucker, 2007), even though in
non-academic contexts this term is sometimes extended to refer to violent
popular uprisings without clear electoral connections.
The main hypotheses to be tested in this section of the book are that
two factors, (H1) the genome of the authoritarian regime and (H2) the way
of its termination, explain the levels of discontinuity between the extinct
authoritarian party systems and those party systems that replace them.
Despite the small number of observations, these hypotheses can be tested
statistically by using a standard tool, multiple linear regression. The depen-
dent variable, substantively defined as party system change, is operational-
ized as the Pedersen index of volatility. It is defined in the same way as in
Chapter 3 of this book. Specifically, I employ the index of volatility between
the last elections in which the given authoritarian party system was manifest,
as listed in Table 2.1 in the column “Most recent election,” and the earliest
b4407_Ch08.indd 200 1/6/2022 9:16:19 AM

elections held after its collapse on a multiparty basis, irrespective of whether

these elections were held in democratic or authoritarian conditions. It has
to be mentioned that as of the time of writing, three countries, Iran, Sudan,
and Yemen, have not held any multiparty elections after the collapse of their
authoritarian party systems. For this obvious reason, they are excluded from
the statistical analysis.
The independent variables and their expected effects on the dependent
variable are thus: The mode of transition is a dichotomous variable coded as
1 for soft transitions, and as 0 for all other modes. I expect that this variable
will be negatively associated with the index of volatility because elite-driven
transitions naturally lead to higher rates of survival of the pre-existing insti-
tutions, which fully applies to party systems. It has to be mentioned that
I experimented with similarly constructed variables referring to military
coups and popular uprisings, but these analyses yielded no statistically
significant results. The results are not reported here because theoretical
reasons for bringing such variables into analysis are not entirely clear. As in
the previous statistical analyses reported in this book, the regime genomes
are also coded as dummy variables. I expect that the genomes of party-based
political regimes, namely, single-party regimes and party electoral author-
itarian regimes, will be negatively associated with the dependent variable.
The effects of non-party electoral authoritarian regimes are not estimated
because only one such observation is available in the dataset.
The small size of the dataset at hand prevents me from engaging many
control variables, but several are necessary. First, I controlled for the effective
number of parties in the last elections in which the given authoritarian party
system was manifest, further in this paragraph referred to as the last elections
for the sake of brevity. It is reasonable to expect that if these elections involve
a large number of opposition parties, which naturally leads to an increase
in party system fragmentation, then a greater number of these parties can
transfer their support bases into the next phase of political development.
Thus, I expect this variable to be negatively associated with party system
change. Second, I control for the level of democracy, operationalized as the
b4407_Ch08.indd 201 1/6/2022 9:16:19 AM

6"×9"
V-Dem index of electoral democracy at the time of the last elections. Since
liberalized dictatorships permit a greater number of opposition parties, some
of which are able to survive regime transition, the impact of this factor upon
the dependent variable is expected to be negative.
If the authoritarian party system is highly volatile, the level of volatility
in the succeeding party system may be also high due to the inertia in the
behavior of voters or party elites. Thus, the third control variable is the level
of volatility in the last elections. It also stands to reason to expect that with
the increase of the time distance between the last elections and the earliest
multiparty elections held after its extinction, the chances for the survival of
individual parties and party loyalties in the electorate become lower. This
explains the inclusion of the related control variable, operationally defined
as the logged time distance between the two aforementioned events in years.
As in all previous statistical analyses reported in this book, all measures of
volatility are defined as absolute shares, not percentage shares. Table 8.2
reports the descriptive characteristics of the non-dichotomous variables.
The results of the analysis are reported in Table 8.3. Note that the numbers
of non-zero values for each of the dichotomous variables are reported in
this table in parentheses.
Of course, the results of this analysis should be met with a great degree
of caution due to the small numbers of observations. In particular, the
Table 8.2. The descriptive characteristics of non-dichotomous variables

in the analysis of post-authoritarian party system change (n = 34).
Standard
Post-authoritarian volatility 0.60 0.60 0.00 1.00 0.29
Index of electoral democracy 0.25 0.25 0.13 0.37 0.06
Effective number of parties in 1.78 1.52 1.00 6.86 1.07
the last elections
Volatility in the last elections 0.19 0.17 0.00 0.52 0.15
Logged time between 0.49 0.39 0.00 1.34 0.36
elections
b4407_Ch08.indd 202 1/6/2022 9:16:19 AM

Table 8.3. Linear regression results, factors of authoritarian party

system change measured as post-authoritarian volatility (n = 34, numbers
of observations in parentheses).
Standard
Soft transition (8) −0.43 0.08 0.00
Monarchy (2) 0.00 0.15 0.98
Military regime (10) −0.23 0.10 0.03
Single-party regime (11) −0.29 0.11 0.01
Post-democracy (2) −0.05 0.10 0.60
Party electoral authoritarian regime (8) −0.21 0.08 0.01
Effective number of parties in the last −0.09 0.04 0.04
elections
Index of electoral democracy −1.48 0.81 0.08
Volatility in the last elections 0.10 0.31 0.74
Logged time between elections −0.03 0.11 0.75
Constant 1.45 0.27 0.00
R-squared = 0.59 — — —
estimates for the regime genomes of monarchy and post-democracy can be

highly unstable. If taken at face value, however, the results strongly support
my main theoretical expectations. In accordance with H1, soft transition
emerges as a highly statistically significant factor and, judging from the value
of the coefficient, as a very strong predictor of the low level of party system
change. Consistent with H2, the genomes of single-party regimes and party
electoral authoritarian regimes are also strongly positively associated with
the dependent variable. It is important that all these associations remain
strong in the presence of the control variables. Two of them, the effective
number of parties in the last elections and the index of electoral democracy,
are important negative predictors of party system change, which is consistent
with my expectations. The two remaining control variables do not exert
statistically significant effects upon party system change, even though the
directions of the observed associations are as expected.
b4407_Ch08.indd 203 1/6/2022 9:16:19 AM

6"×9"
The results of the statistical analysis show that even if the number of
parties in the last elections is controlled for, the genome of military regimes
is negatively associated with post-authoritarian volatility. This finding is
in apparent contradiction with the above theoretical expectation that only
the genomes of party-based political regimes will be negatively associated
with post-authoritarian party system change. Military regimes are certainly
not party-based. Why then the observed association? Of course, it has to
be noticed that this association is not as statistically significant as for the
party-based regime genomes. But it is still strong enough. In this respect,
it can be argued that what we observe here is a different and more complex
pattern of causation.
The analysis presented in Chapter 5 of this book shows that the party
systems associated with military regimes are rather fragmented and highly
volatile. In substantive terms, this combination of party system properties
corresponds to the fact that dominant pro-regime parties under military
regimes are unstable, which in itself may increase the scope of post-
authoritarian regime change. But at the same time, these parties’ links with
the narrow ruling groups of the military regimes are not as strong as it is
the case under the single-party rule. In other words, pro-regime parties
enjoy a higher degree of autonomy, which may facilitate their survival after
regime change. At the same time, higher levels of fragmentation indicate the
presence of relatively important opposition parties that are certainly capable
of outliving military regimes.
The above analysis leads to substantively important results. From a
methodological perspective, the analysis confirms the value of the concept
of regime genomes for our understanding of the dynamics of authoritarian
party systems. This is particularly important at the stage when the focus of
this study shifts to the prospective trends in the development of electoral
authoritarianism. At the same time, the obtained results emphasize the
need to proceed with caution. As demonstrated above, the lines of causation
in the development of authoritarian party systems are not necessarily
straightforward.
b4407_Ch08.indd 204 1/6/2022 9:16:19 AM

The Present and Future of Authoritarian

Party Systems
20 out of 57 authoritarian party systems included in this study are still in
existence. The analysis in this section proceeds in two steps. First, I compare
the properties of the surviving authoritarian party systems with those of the
extinct ones. Second, I identify a set of countries where authoritarian party
systems are likely to be in the making. Both steps converge in the question to
be addressed in this section: what properties of authoritarian party systems
are likely to be observed in the future? Regrettably, the standard methods of
survival analysis, such as the Cox proportional hazards model that has been
applied in a previous study to compare the survival rates of party systems
under electoral authoritarianism, single-party rule, and democracy (Golosov,
2013a), are not applicable here because the amount of the data is small due to
a more restrictive approach to building the empirical sample. This invites an
approach that is more nuanced and, admittedly, less statistically convincing.
The results of the first of these steps in my analysis are presented in
Table 8.4. These results demonstrate that the world of authoritarian party
systems is remarkably stable. None of the differences between the existing
and extinct party systems that can be observed in the table is large, and in
Table 8.4. The properties of authoritarian party systems by survival

status, descriptive statistics.
Standard
Extinct party systems (n = 37)
Fragmentation 1.66 1.51 1.00 5.04 0.73
Volatility 17.52 16.80 0.00 42.17 11.02
Nationalization 0.72 0.72 0.29 0.94 0.18
Existing party systems (n = 20)
Fragmentation 1.99 1.53 1.03 7.92 1.53
Volatility 16.85 17.96 1.56 36.24 9.42
Nationalization 0.71 0.74 0.33 0.98 0.19
b4407_Ch08.indd 205 1/6/2022 9:16:19 AM

6"×9"
the case of one parameter, nationalization, the difference is literally absent.

The results of a more rigorous comparison of the means, not reported here,
have demonstrated that none of the differences is statistically significant. It is
of course noticeable that the mean effective number of parties of the extinct
systems is smaller than that of the surviving ones, but the comparison of
the median values and standard deviations immediately reveals the reason
why this difference is not statistically significant. Figure 8.1 illustrates this
by showing the mean and median values of fragmentation of authoritarian
party systems on a year-to-year basis, so that each of the data points registers
the mean (Plot 1) or median (Plot 2) fragmentation of the party systems
in place in the given year. The lack of dynamics demonstrated by Plot 2 is
particularly striking.
In order to understand this remarkable stability of the main parameters
of authoritarian party systems, it is important to cast a closer look at the dis-
tributions of the extinct and surviving systems by those categories that have
been found to be important in the previous chapters of this book, political
regime genomes and types of executive–legislative relations. The results of
this analysis are presented in Table 8.5. The ANOVA comparisons, reported
in the table only by indicating statistical significance, yield a significant result
for only one regime genome, the non-party electoral authoritarian regime.
This regime genome is nearly absent among the extinct party s ystems but
rather widespread, second only to single-party regimes, among the exist-
ing ones. But, given the small size of the sample and the large number of
categories, it does not make sense to ignore statistically insignificant dif-
ferences, particularly a great decline in the share of military regimes, and
great increases in the shares of single-party regimes and semi-presidential
systems. None of the systems with parliamentary elected executive presidents
remains in place.
The analysis presented in Chapter 5 of the book makes it possible to
suggest that the observed lack of differences between the extinct and sur-
viving authoritarian party systems can be parsimoniously explained with
b4407_Ch08.indd 206 1/6/2022 9:16:19 AM

2.5
Plate 1
1.5
0.5
0
1949 1959 1969 1979 1989 1999 2009 2019
2.5
Plate 2
1.5
0.5
0
1949 1959 1969 1979 1989 1999 2009 2019
Figure 8.1. Mean and median fragmentation of authoritarian party sys-

tems by year, 1949–2019. Plate 1: Mean. Plate 2: Median.
b4407_Ch08.indd 207 1/6/2022 9:16:19 AM

6"×9"
reference to the composition of these two party system categories by regime

genomes. The shares of two of them, single-party regimes and non-party
electoral authoritarian regimes, are about 20 percent greater among the
existing systems than among the extinct systems, while the share of military
regimes declined by more than 15 percent. My previous analysis has estab-
lished that the genome of military regimes is strongly positively associated
with party system fragmentation and volatility; the reverse — a strongly
negative association — is true of the genome of single-party regimes; and
the effects of non-party electoral authoritarian regimes on both parame-
ters are neutral. Thus, the decline in the share of military regimes has been
compensated by the rise in the share of single-party regimes to the effect
that on the average, the parameters of authoritarian party systems remain
unchanged.
Table 8.5. Distribution of the extinct and existing authoritarian party

systems by political regime genomes and institutional design types (%).
Extinct Existing
party party
systems, systems, Significance
n = 37 n = 20 (ANOVA)
Political regime genomes
Monarchy 8.11 5.00 0.67
Military regime 32.43 15.00 0.16
Post-democracy 5.41 5.00 0.95
Party electoral authoritarian regime 21.62 15.00 0.55
Non-party electoral authoritarian regime 2.70 20.00 0.03
Institutional design types
Parliamentary 8.11 15.00 0.43
System with parliamentary elected 10.81 0.00 0.13
b4407_Ch08.indd 208 1/6/2022 9:16:19 AM

The difference between the compositions of the extinct and surviving

authoritarian party systems by institutional design types is less pronounced.
However, it is possible to argue that a similar process of compensation occurs.
Semi-presidential systems support both party system fragmentation and
volatility, but a great increase in the share of these systems has occurred in
parallel with the rise in the share of parliamentary regimes with their oppo-
site effects on both parameters. The picture is far less clear than in the case
of political regime genomes, though. It has to be emphasized that with the
data at hand, the reasoning above, be it with regard to regime genomes or
institutional design types, cannot be sufficiently supported by quantitative
analysis. Nevertheless, the observed regularities can be taken into account
in a modest attempt of prognostication that is to follow.
The universe of authoritarian party systems may be stable within its own
boundaries, but the boundaries are ever changing and fuzzy. Of course, the
number of authoritarian regimes in the contemporary world is far greater
than 20. Quite a few of such regimes are non-electoral, as is the case with the
remaining absolute monarchies and single-party regimes. Others persistently
conduct elections on a non-party basis. It is remarkable to note that even
after the introduction of a mixed electoral system in Jordan, the legislature
of the country is still dominated by pro-regime independents (Singh, 2017).
Egypt, after several decades of multiparty politics, descended into a similarly
non-partisan pattern of competition after the 2013 military coup (Völkel,
2017). Nevertheless, the instances of party-structured authoritarian elections
are numerous enough. But relatively few countries can be considered as
developing stable patterns of electoral politics that lead to the emergence of
continuous authoritarian party systems. For example, elections conducted in
Myanmar in 2010 and 2015 were held in different political conditions and
with different sets of important parties (Barany, 2018), and the same can be
said about the 2015 and 2020 elections in Venezuela.
I was able to identify seven countries that, by meeting all but one crite-
rion set in Chapter 2 of this book, could be qualified as developing author-
itarian party systems as of the end of 2019, and where these processes were
b4407_Ch08.indd 209 1/6/2022 9:16:19 AM

6"×9"
not terminated by coups or democratizations as of April, 2021. The unmet

criterion is the number of elections in which the prospective authoritarian
party systems have been made manifest. This number has been set as three
for the full-fledged systems that have been so far analyzed in this study, but
it is lower for the nascent systems. Table 8.6 provides descriptive information
about these country cases. It has to be recognized that the case of Thailand
is particularly problematic because, due to the fact that the current political
regime of the country originated from the 2014 military coup, it could be
alternatively characterized as military by regime genome and parliamen-
tary by institutional regime type. But, taking into account that the defense
of the monarchy was not only proclaimed as the main goal of the coup by
its leaders but also practically implemented in the process of post-coup
institution building (Chambers and Waitoolkiat, 2016), I found it possible
to characterize Thailand as a monarchy on both parameters.
There is no reason to expect that all these nascent authoritarian party
systems will reach maturity. Bangladesh and Pakistan, in particular, have
already experienced fluctuations between authoritarianism and democracy,
and such fluctuations may occur in the future as well. Thus, any quanti-
tative assessments derived from this data are preliminary at best. Still, it
is possible to make two substantively important observations. First, the
effective number of parties in these systems, 2.58 on average, is much
higher than in the authoritarian party systems analyzed in this book. It
has to be kept in mind that, as demonstrated by the previously reported
statistical analyses, the number of parties normally decreases in the second
elections in which an authoritarian party system is manifest. Given that
most of the nascent systems did not reach this phase, this partly explains
the observed difference, but the discrepancy is too great to be explained
by this factor alone. Note that in both countries that have held series of
two elections, Bangladesh and Turkmenistan, the levels of party system
fragmentation did decrease over time, but in the latter case, fragmentation
remains high enough.
b4407_Ch08.indd 210 1/6/2022 9:16:19 AM

Table 8.6. The emerging authoritarian party systems as of the end of

2019.
Effective
Year of number
most of parties
recent (mean
national if more
legislative Number of Institutional than one
Country elections elections Regime genome design type election)
Bangladesh 2014 2 Post-democracy Parliamentary 1.22
Guinea 2013 1 Military Presidential 2.58
Nicaragua 2016 1 Post-democracy Presidential 1.30
Pakistan 2018 1 Post-democracy Parliamentary 2.70
Thailand 2019 1 Monarchy Monarchy 5.42
Turkey 2018 1 Post-democracy Presidential 2.39
Turkmenistan 2018 2 Single-party Presidential 2.48
An additional explanation can be derived from the fact that one regime
genome, post-democracy, while limited to just three observations among
the 57 party systems included in the main analysis, comprises more than
a half of cases among the nascent authoritarian party systems. The main
reason for this situation is the global tendency described in the literature
under the label of autocratization (Maerz et al., 2020), which is defined as
the “decline of democratic regime attributes” (Lührmann and Lindberg,
2019, p. 1095). This simple definition clearly implies that the countries
experiencing autocratization do have currently, or used to have in the near
past, the necessary attributes for democracy. Thus, the connection between
the trend toward autocratization and the rising number of post-democracies
among the emerging authoritarian regimes is self-explanatory.
It is important to emphasize that autocratization is a complex overarch-
ing concept (Cassani and Tomini, 2019). It is not necessarily equivalent to
the complete breakdown of democracy. Some of the countries experiencing
decline of democratic regime attributes still retain these attributes to an
b4407_Ch08.indd 211 1/6/2022 9:16:19 AM

6"×9"
extent that is sufficient for them not to qualify as authoritarian regimes. But
in other countries, as illustrated by the data in Table 8.6, autocratization has
progressed far enough to result in full-fledged authoritarianism. Thus, there
are reasons to expect that in the future the share of post-democracies among
authoritarian regimes will increase, which makes it possible to develop an
informed speculation about the properties of the future authoritarian party
systems. Unfortunately, the empirical foundations for such speculation, as
laid down in this study, are shaky because post-democracy is a rare regime
genome among the extinct and existing party systems explored in this
study. However, certain very tentative conclusions can be drawn from the
obtained results, as well as from the theoretical reasoning that has been used
to explain these results.
It seems that post-democracy is likely to generate party systems that
are neither highly fragmented nor highly concentrated. On the one hand,
high levels of fragmentation are effectively prohibited by the fact that elec-
tions provide the only mode of power acquisition available to such political
regimes, as a result of which they cannot afford losing elections. Of course,
an increase in the effective number of parties makes this threat more
credible, which pushes the autocrats toward taking pre-emptive measures.
On the other hand, the party legacies of post-democracies are richer than
those of other political regime genomes. This suggests that the number of
survivors among the parties inherited from democracy can be large enough,
particularly because some of these parties can be correctly estimated by the
regime as posing no threat to autocracy because they have demonstrated
their impotence in preventing it.
At the same time, the autocratic elites of post-democracies acquire their
political skills in democratic conditions, thus learning how to neutralize
the threats posed by multipartism without uprooting any semblance of
opposition. Moreover, these elites, once propelled to political dominance
by democratic procedures, value the attributes of democracy — including
the existence of loyal party-structured opposition — as their own political
assets. Post-democratic authoritarian regimes claim to be democracies even
b4407_Ch08.indd 212 1/6/2022 9:16:19 AM

at the risk of depriving themselves of the alternative bases of ideological

legitimacy (Carothers, 2018).
The analysis performed in the previous chapters of this book allows for
an informed speculation that the party systems of the post-democracies will
be volatile. To a greater extent than previously widespread political regime
genomes, post-democracies can be expected to display a tendency toward
personalization of power. This tendency follows from the fact that in the
absence of democracy, programmatic competition among political parties
becomes secondary to competition based on the perceived personal qualities
of individual politicians, which increases the leverage of the incumbent rul-
ers. Post-democratic autocrats widely employ populist rhetoric (Stockemer,
2019). They tend to avoid close associations with those parties that support
them, as is quite evident from the experiences of an established post-
democratic regime, Russia (Robinson and Milne, 2017). Even if such asso-
ciations are historically strong, pro-regime parties tend to be sidelined by
their leaders both in government and in the electoral arenas, as it happened
in Turkey in the process of authoritarian transformation (Lancaster, 2014).
Thus, in post-democracies, the stability of dominant pro-regime parties
is not a paramount priority for the authoritarian executives. This serves as a
source of party system volatility. The second source is directly related to the
fact that one of the ways of ensuring electoral victories commonly employed
by post-democratic regimes is the continuous reshuffling of the opposition.
This is achieved not only by direct party bans (Sinpeng, 2014) but also, and
increasingly, by manipulating the field of party alternatives in a way that
prevents the development of stable party loyalties among the anti-regime
voters (Tertytchnaya, 2020; Turchenko and Golosov, 2021). Similar purposes
can be achieved by promoting decoy parties that enter the electoral arenas
either for splitting the opposition vote (Golosov, 2015a) or for creating the
visibility of cross-party support for the authoritarian executive (Pavlović,
2020). Besides, authoritarian regimes can selectively offer limited support
to some opposition parties and repress or discredit others (Nugent, 2020),
which not only increases polarization within the opposition camp but also
b4407_Ch08.indd 213 1/6/2022 9:16:19 AM

6"×9"
affects the behavior of critically minded voters, thus facilitating party system
volatility.
Therefore, it can be cautiously posited that the authoritarian party sys-
tems of the future will be rather, but not excessively, fragmented, and that
they will be relatively volatile. In other words, they will be quite similar to
the party systems of contemporary democracies. Thus, authoritarianism
will emulate democracy more plausibly than ever before. This cautious
prognosis can be juxtaposed to the main finding of a recent comparative
study of Bernhard et al. (2020) who show that for electoral authoritarian
regimes, higher levels of electoral competition are most dangerous in the
short term during the first three elections conducted under these regimes.
Once the threshold of three consequent elections is passed, the survival of
the regime crucially depends on its ability to reduce uncertainty over the
outcomes of multiparty elections. The empirical perspective of my study is
different because it deals exclusively with those regimes that have already
passed the threshold. Is it possible to expect that the new regimes, assuming
that they will arrive mostly in the form of post-democracies, will be able to
cope with the challenge of multiparty elections?
Bernhard et al. (2020, p. 481) tentatively conclude that “unless they
can devise effective strategies for mastering electoral competition, dicta-
tors who stake their future on democratic emulation are more like gam-
blers than contemporary Machiavellis.” Hence, the future of authoritarian
regimes hinges on their ability to keep electoral competition under control.
I would suggest that the projected properties of the party systems of post-
democracies make the gambling metaphor more plausible than the
Machiavellian one. If the strength of the authoritarian executive remains
unshaken, so that the outcome of the gamble is certain, then keeping a
rather fragmented and volatile pattern of party competition under control
can be achieved with ease, particularly if such a strategy is facilitated by the
previous democratic experience of the ruling group. Besides, this pattern is
optimal for emulating democracy and, for this reason alone, it retains some
value for the autocrats.
b4407_Ch08.indd 214 1/6/2022 9:16:19 AM

If, however, the regime goes through hard times, be it because of

the succession crisis, economic difficulties, or external shocks, then
gambling becomes risky. In times of crisis, the projected properties
of post-democratic party systems are not very promising for regime
survival. From the perspective of democratization, this prediction is not as
optimistic as it may seem. Learning from failure may push dictators into a
more Machiavellian mode of operation, as a result of which the projected
properties of the post-democratic party systems will wither away. The
parameters of authoritarian party systems have been stable so far, and they
may retain stability by uprooting the democratic legacies of the emerging
dictatorships.
b4407_Ch08.indd 215 1/6/2022 9:16:19 AM

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Conclusion
I
nstitutions have always been important for autocracies. Those of them
that relied mostly on violence and arbitrary rule did not last long. In
order to achieve long-term survival, many autocracies of the past devel-
oped institutional arrangements that were essential for selecting political
leaders, for maintaining them in power, and therefore, for the succession
of power. Such arrangements constitute political regimes. As long ago as in
the 19th century, some of the autocracies started to use elections as second-
ary components of their institutional structures. The presence of elections
did not alter the fundamental characteristics of these regimes. As long as
authoritarianism persisted, the electoral expression of the will of the peo-
ple could not prevail over the intrinsic mechanisms of power acquisition
and succession that were characteristic of the autocracies of the past and
their main categories: monarchies, military regimes, single-party regimes,
and personal dictatorships. But none of them completely avoided holding
quasi-competitive elections. This option was always on the table.
Thus, in retrospect, electoral authoritarianism is not a new phenome-
non. What is new is that in the recent decades, and especially after the fall
of global communism in the end of the 1980s, electoral authoritarianism
emerged as the most widespread form of autocracy. In part, this happened
because monarchies, military regimes, and single-party regimes have become
less numerous than in the past, while personal dictatorships flourish. Due
to the weakness, if not to say absence, of the regime-specific mechanisms
of power acquisition and succession, personalist dictatorships are partic-
ularly inclined to develop institutional structures that include elections as
217
b4407_Conclusion.indd 217 1/6/2022 9:16:31 AM

6"×9"
their essential component. The role of elections remains secondary, but it is

certainly more pronounced than in the past. Contemporary autocracies are
widely engaged in democratic emulation. They wish to look as democratic
as they can afford to without falling into the risk of losing power as a result
of elections. As long as the autocrats’ grip on power is firm, the threat of
democratization by elections can be neutralized by using a vast repertoire
of manipulations at the disposal of the regime.
As a result, contemporary autocracies not only hold elections, but they
also tend to hold them on a multiparty basis. This leads to the emergence
and global spread of authoritarian party systems. While the individual com-
ponents of authoritarian party systems — such as the dominant pro-regime
parties — have been subjected to investigation in the recent literature, little
research has been done so far on these systems’ properties. These proper-
ties, including fragmentation, format, volatility, and nationalization, can
be observed in all party systems, democratic and authoritarian alike, but so
far, a systemic perspective on authoritarian party politics has been avoided
in the literature. This book fills the gap by providing an inquiry into the
properties of authoritarian party systems.
The theoretical considerations presented in the Chapter 2 of the book
make it possible to delineate the universe of contemporary authoritarian
party systems by identifying 57 country-level cases that comprise 294 elec-
tions held in 55 countries from 1945–2019. Chapter 3 provides essential
factual information about these cases and their historical antecedents. The
analysis reveals that while authoritarian party systems have always been
around, their number has markedly increased over the duration of the
post-World War II period. The spatial shift is also apparent. After the third
wave of democratization, a majority of authoritarian party systems can be
found in Africa and in the post-Communist world, but in the course of the
previous period, they were mostly located in the West Hemisphere and East
Asia. Despite the observed dynamics, the past and present systems display
significant similarities in the modes of operation. Variations among them
result from structural rather than temporal or spatial factors.

Conclusion 219
In order to establish the sources of these variations by means of quan-

titative analysis, it is essential to register them with numerical precision.
Chapter 4 discusses the methodological tools of measuring party systems’
properties and, where necessary, refines these methods in order to apply them
more efficiently to the study of authoritarian party systems. This opens the
possibility of drawing systematic comparisons between the party systems of
democracies and autocracies. Quite expectedly, the analysis demonstrates
that on average, authoritarian party systems are decisively less fragmented
than those of democracies. This differentiates authoritarian party systems
not only from the party systems of well-established democracies but also
from those of more uncertain democratic states. High levels of party frag-
mentation are not alien to authoritarianism, but as a tendency, autocracies
generate predominant party systems. The levels of volatility of authoritarian
party systems are about the same as in the countries at the apex of democratic
development, but decisively lower than in other democracies. However, with
respect to party system nationalization, a contrasting pattern is evident: the
party systems of autocracies are about as nationalized as in less than perfect
democracies, while the party systems of well-established democracies tend
to display higher levels of nationalization.
Chapter 5 investigates the variations among authoritarian party systems
from the perspective of the political regimes with which they are associ-
ated. In order to perform this task more efficiently, the chapter introduces
the concept of political regime genome as an integral characteristic of the
regimes’ origin and mode of operation, and identifies six regime genomes
as the main analytical tool for further analysis: monarchy, military regime,
single-party regime, post-democracy, party electoral authoritarian regime,
and non-party electoral authoritarian regime. The last three categories
represent major varieties of contemporary personalist dictatorships. The
theoretical argument presented in this chapter posits that the variable effects
of the regime genomes upon party system properties can be related to two
main aspects, the mode of power acquisition and party legacies. Both aspects
are inherent in the substantive definitions of regime genomes.

6"×9"
The analysis shows that monarchies and military regimes stand out
as genomes that are associated with most fragmented party systems. The
reverse is true with regards to the genomes of single-party regimes and party
electoral authoritarian regimes. Among the predominant party systems,
high fragmentation within the opposition camp is not characteristic of
military regimes, while the genomes of single-party systems and especially
non-party electoral authoritarian regimes tend to be associated with
fragmented oppositions. The party systems of monarchies, on the one
hand, and military regimes and post-democracies, on the other hand, are
volatile. In this respect, they are significantly different from the party systems
of single-party regimes and party electoral authoritarian regimes, but not
from those of non-party electoral authoritarian regimes. On the parameter
of party system nationalization, monarchies score low, which contrasts them
to the genome of party electoral authoritarian regimes.
Chapter 6 investigates the impact of the formal institutional structures,
models of executive–legislative relations (parliamentary systems, systems
with parliamentary elected executive presidents, presidential, and semi-
presidential systems) and federalism, upon the properties of authoritarian
party systems. The institutions dealt with in this chapter are highly endoge-
nous, which does not invalidate institutional explanations per se but makes it
imperative to pay primary attention to the underlying structures of political
incentives. These structures are theoretically explained with reference to
varying degrees of political risk-taking affordable for the autocrats under
different institutional arrangements, and to their ability to avoid the related
dangers by hedging the risks, i.e., by using alternative ways of preserving
power that are allowed by the existing institutional arrangements.
The analysis reveals that the party systems of authoritarian parliamen-
tary regimes are significantly less fragmented than those that exist under
different institutional arrangements. Semi-presidential systems are associated
with most fragmented party systems, followed by the systems with parlia-
mentary elected executive presidents and presidential systems, and distantly
traced by parliamentarism. Parliamentary and presidential systems are

Conclusion 221
disposed toward predominant party systems with concentrated opposition,

while semi-presidentialism is associated with high fragmentation within
the opposition camp. In terms of volatility, the party systems of parliamen-
tary regimes and systems with parliamentary elected executive presidents
are more stable than those in all other categories. The semi-presidential
authoritarian party systems demonstrate a particularly high level of vola-
tility. The analysis of party system nationalization shows that systems with
parliamentary elected executive presidents are associated with the highest
levels of party system nationalization, while the level of nationalization of
the authoritarian semi-presidential party systems tends to be low. The party
systems of authoritarian federations tend to be less fragmented than those
of unitary states, but the effect is not very pronounced in the analysis of
fragmentation as such and fades away in the analysis of party system format.
With a greater degree of certainty, it can be said that authoritarian federal-
ism is associated with fragmented oppositions. The effects of authoritarian
federalism upon party system volatility and nationalization are insignificant.
Chapter 7 is focused on electoral system effects. The descriptive anal-
ysis presented at the beginning of the chapter shows that in comparison to
democracies, autocracies have a strong preference towards majoritarian
systems (including their varieties that are rare in contemporary democ-
racies, such as multimember plurality and party block vote), and a very
well-articulated preference toward mixed systems. Indeed, some of the
most complex and idiosyncratic mixed electoral systems of the world can
be found in authoritarian regimes. Theoretically, the autocrats’ preferences
can be related to the incentive structures of risk aversion and risks hedging
that have been identified in the Chapter 8. The reasons why the autocrats
sometimes opt for proportional representation or mixed rules are often
connected to their desire to facilitate the organizational development of
dominant parties. Institutional legacies and territorial diffusion also play
a role. The analysis demonstrates that once the electoral system of the
autocrats’ choice is in operation, the effects of electoral rules are similar
to those registered in research on democracies. In comparison to other

6"×9"
electoral rules, proportional representation leads to more fragmented, more

volatile, and more nationalized party systems. The effects of majoritarian
rules are opposite. The analysis shows that in authoritarian conditions, the
elements of proportional representation in mixed electoral systems are
inconsequential for party system properties. The effects of mixed systems
are hardly distinguishable from those of purely majoritarian rules.
Chapter 8 discusses the dynamic aspects of authoritarian party
development by focusing, first, on the legacies they leave behind after they
cease to exist, and second, on the prospects for the existing ones and those
yet to come. Speaking of the former aspect, the analysis demonstrates that
two factors, the genome of the authoritarian regime and the way of its
termination, explain the levels of discontinuity between the extinct authori-
tarian party systems and those party systems that replace them. The analysis
shows that soft transition, understood as a regime transition in which the
ruling group of the incumbent authoritarian regime plays a significant role,
emerges as a strong predictor of the low level of party system change after
the collapse of authoritarian regime. Among the regime genomes, military
regimes, single-party regimes, and party electoral authoritarian regimes
are associated with higher rates of survival of the residua of authoritarian
party systems.
The comparison of the parameters of the extinct and existing authoritar-
ian party systems shows that the properties of these systems are remarkably
stable, to the effect that in both categories, the levels of fragmentation are very
similar. This can be tentatively explained with reference to the composition
of these two party system categories by regime genomes. From this perspec-
tive, it is important to take into account the trend toward autocratization
of political regimes that is observable in the contemporary world. This trend,
resulting in the rising number of post-democracies among the emerging
authoritarian regimes, may suggest that the authoritarian party systems of
the future will be rather, but not excessively, fragmented, and that they will be
relatively volatile. The projected properties of post-democratic party systems
are not very promising for the survival of authoritarian regimes. But there

Conclusion 223
are no grounds for excessive optimism in our estimates of the prospects for
transition to democracy. The parameters of authoritarian party systems have
been stable so far, and they may retain stability if the autocrats, with their
ability to learn from failure, will be persistent in uprooting the democratic
legacies of the emerging dictatorships.

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

List of the Names of Parties

and Political Organizations
in Native Languages
Party name in English Country Party name in native language

All-Russia People’s Front Russia Obshcherossiyskii narodnyi front
Arab Socialist Ba’ath Party Arab countries Hizb al-Ba‘th al-’Arabıˉ al-Ishtira
ˉkıˉ
˙
Brazilian Democratic Movement Brazil Movimento Democrático Brasileiro
Cambodian People’s Party Cambodia Kanakpak Pracheachon Kampuchea
Cameroon People’s Democratic Cameroon Rassemblement démocratique du
Movement Peuple Camerounais
Catholic Party Belgium Dutch: Katholieke Partij; French:
Parti catholique
Christian Democratic Party El Salvador Partido Demócrata Cristiano
Civic Party of Kazakhstan Kazakhstan Qazaqstan Azamattyq Partııasy
Colombian Conservative Party Colombia Partido Conservador Colombiano
Colombian Liberal Party Colombia Partido Liberal Colombiano
Communist Party of the Russia Kommunisticheskaya Partiya
Russian Federation Rossiiskoi Federatsii
Congolese Party of Labour Republic of Parti congolais du travail
Congo
Democratic Action Party Malaysia Malay: Parti Tindakan Demokratik;
Chinese: Mínzhǔ Xíngdòng Dǎng
Democratic and Social Mauritania Parti républicain et démocratique
Republican Party social
Democratic Constitutional Rally Tunisia At-Tajammu’ ad-Dustūrıˉ
ad-Dıˉmuqraˉtˉı
˙
Democratic Republican Party South Vietnam Đảng Cộng hòa Dân chủ
Ethiopian People’s Ethiopia Ye’ıˉtiyop’iya ḥizibochi a
ˉbiyotawıˉ
Revolutionary Democratic Front dıˉmokirasıˉyawıˉ ginibari
(Continued)
225
b4407_List of the Names.indd 225 1/6/2022 9:17:12 AM

6"×9"

General People’s Congress Yemen Al-Mo’tamar Ash-Sha’abiy Al-’Aam
German Centre Party Germany Deutsche Zentrumspartei
Indonesian Democratic Party Indonesia Partai Demokrasi Indonesia
Institutional Democratic Party Guatemala Partido Institucional Democrático
Institutional Revolutionary Party Mexico Partido Revolucionario Institucional
Islamic Renaissance Party of Tajikistan Hizbi Nahzati Islomii Tojikiston
Tajikistan
The Left Denmark Venstre
Liberal Democratic Party of Russia Liberal’no-Demokraticheskaya
Russia Partiya Rossii
Liberal Front Party Brazil Partido da Frente Liberal
Liberal Party Belgium Dutch: Liberale Partij; French: Parti
liberal
Liberal Party of the Philippines Philippines Partido Liberal ng Pilipinas
Malaysian United Indigenous Malaysia Parti Pribumi Bersatu Malaysia
Party
Mexican Communist Party Mexico Partido Comunista Mexicano
Nationalist (Nacionalista) Party Philippines Partido Nacionalista
National Action Party Mexico Partido Acción Nacional
National Conciliation Party El Salvador Partido de Conciliación Nacional
National Democratic Party Egypt Al-Hizb Al-Watanıˉ Ad-Dıˉmūqraˉtˉı
˙ ˙ ˙
National Democratic Party Nepal Rastriya Prajatantra Party
National Democratic Rally Algeria At-tajammu‘ al-waţanıˉ ad-dıˉmūqraˉţıˉ
National Front Malaysia Barisan Nasional
National Front of Iran Iran Jebhe-ye Melli-ye Ira
ˉn
National Liberation Front Algeria Jabhatu l-Taḥrıˉri l-Watanıˉ
˙
National Liberation Movement Guatemala Movimiento de Liberación Nacional
National Opposition Union Nicaragua Unión Nacional Opositora
National Party South Africa Afrikaans: Nasionale Party
National Renewal Alliance Brazil Aliança Renovadora Nacional
National Republican Paraguay Asociación Nacional Republicana —
Association — Colorado Party Partido Colorado
National Revolutionary South Vietnam Phong trào cách mạng quốc gia
Movement
National Union Portugal União Nacional
National Union for Democracy Cameroon Union Nationale pour la Démocratie
and Progress et le Progrès

List of the Names of Parties and Political Organizations 227

National Unity Party Hungary Nemzeti Egység Pártja
Nationalist Liberal Party Nicaragua Partido Liberal Nacionalista
New Azerbaijan Party Azerbaijan Yeni Azərbaycan Partiyası
New Iran Party Iran Hezb-i Īraˉn-i Nuvıˉn
˙
Party of the Independence Madagascar Malagasy: Antoko’ny Kongresi’ny
Congress of Madagascar Fahaleovantenan’i Madagasikara;
French: Parti du Congrès de l’in-
dépendence de Madagascar
Patriotic Salvation Movement Chad Mouvement patriotique du salut
People’s Action Party Singapore Chinese: Rénmín Xíngdòngdǎng;
Malay: Parti Tindakan Rakyat
People’s Democratic Party of Tajikistan Hizbi xalqii demokratii Tojikiston
Tajikistan
People’s Democratic Party of Uzbekistan O’zbekistan Xalq Demokratik
Uzbekistan Partiyasi
People’s Party Iran Hezb-e Mardom
˙
People’s Party for Democratic Parti du Peuple pour la
Reconstruction and Democracy Republic of the Reconstruction et la Démocratie
Congo
People’s Rally for Progress Djibouti Rassemblement populaire pour le
Progrès
Reformist Party Dominican Partido Reformista
Republic
Republican Party of Armenia Armenia Hayastani Hanrapetakan
Kusaktsutyun
Revolutionary Nationalist Bolivia Movimiento Nacionalista
Movement Revolucionario
Revolutionary Party Guatemala Partido Revolucionario
The Right Denmark Højre
Serbian Radical Party Serbia Srpska radikalna stranka
Social Democratic Party Madagascar Parti Social Démocrate
Social Democratic Party of Germany Sozialdemokratische Partei
Germany Deutschlands
Socialist Party of Serbia Serbia Socijalistička partija Srbije
Socialist Union of Popular Morocco Al-Ittihad Al-Ishtirakiy Lilqawat
Forces Al-Sha’abiyah
Traditionalist Spanish Phalanx Spain Falange Española Tradicionalista y
and (that) of the Councils de las Juntas de Ofensiva Nacional
of the National Syndicalist Sindicalista
Offensive
(Continued)

6"×9"

United Development Party Indonesia Partai Persatuan Pembangunan
United Malays National Malaysia Pertubuhan Kebangsaan Melayu
Organization Bersatu
United Russia Russia Yedinaya Rossiya
Uzbekistan Liberal Democratic Uzbekistan O‘zbekiston Liberal Demokratik
Party Partiyasi
Yemeni Congregation for Yemen Al-Tajummu’ al-Yamani lil-islah
Reform
Yemeni Socialist Party Yemen Al-Hizb al-Ishtiraki al-Yamani

References
Abbink, J. (2011). Ethnic-based federalism and ethnicity in Ethiopia: Reassessing

the experiment after 20 years. Journal of Eastern African Studies, 5(4), 596–618.
Aburamoto, M. (2019). The politics of anti-corruption campaigns in Putin’s Russia:
Power, opposition, and the All-Russia People’s Front. Europe-Asia Studies,
71(3), 408–425.
Acemoglu, D. and Robinson, J. A. (2006). Economic Origins of Dictatorship and
Democracy. Cambridge, UK: Cambridge University Press.
Acemoglu, D., Egorov, G., and Sonin, K. (2008). Coalition formation in non-
democracies. Review of Economic Studies, 75(4), 987–1009.
Adamson, F. B. (2020). Non-state authoritarianism and diaspora politics. Global
Networks, 20(1), 150–169.
Ahmed, A. (2010). Reading history forward: The origins of electoral systems in
European democracies. Comparative Political Studies, 43(8–9), 1059–1088.
Al-Awadi, H. (2009). A struggle for legitimacy: The Muslim Brotherhood and
Mubarak, 1982–2009. Contemporary Arab Affairs, 2(2), 214–228.
Aldrich, J. H. (1995). Why Parties? The Origin and Transformation of Political Parties
in America. Chicago, IL: University of Chicago Press.
Alexander, J. and McGregor, J. (2013). Introduction: Politics, patronage and violence
in Zimbabwe. Journal of Southern African Studies, 39(4), 749–763.
Alexander, R. J. (Ed.) (1982). Political Parties of the Americas. Westport, CT:
Greenwood.
Allinson, M. (2000). Politics and Popular Opinion in East Germany, 1945–68.
Manchester, UK: Manchester University Press.
Amjad, M. (1989). Iran: From Royal Dictatorship to Theocracy. Westport, CT:
Greenwood.
Amorim Neto, O. and Cox, G. W. (1997). Electoral institutions, cleavage structures
and the number of parties. American Journal of Political Science, 41(1), 149–174.
Anckar, C. (2000). Size and party system fragmentation. Party Politics, 6(3), 305–328.
229
b4407_References.indd 229 1/6/2022 9:17:23 AM

6"×9"
Anckar, C. (2020). Constitutional monarchies and semiconstitutional monarchies:

A global historical study, 1800–2017. Contemporary Politics, 27, 23–40.
Anckar, D. and Anckar, C. (2000). Democracies without parties. Comparative
Political Studies, 33(2), 225–247.
Anderson, M. L. (2000). Practicing Democracy: Elections and Political Culture in
Imperial Germany. Princeton, NJ: Princeton University Press.
Anh, N. T. (1985). The Vietnamese monarchy under French colonial rule, 1884–1945.
Modern Asian Studies, 19(1), 147–162.
Ash, K. (2015). The election trap: The cycle of post-electoral repression and
opposition fragmentation in Lukashenko’s Belarus. Democratization, 22(6),
1030–1053.
Azimi, F. (1997). On shaky ground: Concerning the absence or weakness of political
parties in Iran. Iranian Studies, 30(1–2), 53–75.
Baghdasaryan, M. (2017). The practice of political rights and patron-client relations:
A case study of a party in Armenia. Citizenship Studies, 21(8), 1034–1051.
Barany, Z. (2018). Elections and constitutional constraints: How the generals have
stayed in power in Myanmar. SAIS Review of International Affairs, 38(1),
105–117.
Barnea, S. and Rahat, G. (2007). Reforming candidate selection methods: A three-
level approach. Party Politics, 13(3), 375–394.
Barracca, S. (2007). Gubernatorial politics and the evolution toward democratic
federalism in Mexico. Regional & Federal Studies, 17(2), 173–193.
Bartolini, S. and Mair, P. (1990). Identity, Competition, and Electoral Availability: The
Stabilisation of European Electorates 1885–1995. Cambridge, UK: Cambridge
University Press.
Basedau, M. and Stroh, A. (2011). Do party systems make democracy work? A com-
parative test of party-system characteristics and democratization in Francophone
Africa. In: Erdmann, G. and Kneue, M. (Eds.), Regression of Democracy?
pp. 173–199. Wiesbaden, Germany: VS Verlag für Sozialwissenschaften.
Baturo, A., and Elgie, R. (2018). Why do authoritarian regimes adopt bicameralism?
Cooptation, control, and masking controversial reforms. Democratization,
25(5), 919–937.
Baturo, A., and Elgie, R. (Eds.) (2019). The Politics of Presidential Term Limits.
Oxford, UK: Oxford University Press.
Batz, G. (2013). Military factionalism and the consolidation of power in 1960s
Guatemala. In: V. Garrard-Burnett, M. A. Lawrence, and J. E. Moreno (Eds.),

References 231
Beyond the Eagle’s Shadow: New Histories of Latin America’s Cold War, pp. 64–65.
Albuquerque, NM: University of New Mexico Press.
Baylis, T. A. (1998). Elite change after communism: Eastern Germany, the Czech
Republic, and Slovakia. East European Politics and Societies, 12(2), 265–299.
Bazenguissa-Ganga, R. (1999). The spread of political violence in Congo-Brazzaville.
African Affairs, 98(390), 37–54.
Beaulieu, E. and Hyde, S. D. (2009). In the shadow of democracy promotion: Strategic
manipulation, international observers, and election boycotts. Comparative
Bedford, S. (2017). ‘The election game’: Authoritarian consolidation processes in Belarus.
Demokratizatsiya: The Journal of Post-Soviet Democratization, 25(4), 381–405.
Benoit, K. (2000). Which electoral formula is the most proportional? A new look
with new evidence. Political Analysis, 8(4), 381–388.
Benoit, K. (2001). District magnitude, electoral formula, and the number of parties.
European Journal of Political Research, 39(2), 203–224.
Bernhard, M., Edgell, A. B., and Lindberg, S. I. (2020). Institutionalizing electoral
uncertainty and authoritarian regime survival. European Journal of Political
Research, 59(2), 465–487.
Bernhard, M., Nordstrom, T., and Reenock, C. (2001). Economic performance,
institutional intermediation, and democratic survival. Journal of Politics, 63(3),
775–803.
Berwouts, K. and Reyntjens, F. (2019). The Democratic Republic of Congo: The
great electoral robbery (and how and why Kabila got away with it). Africa
Policy Brief. No. 25, 19 April.
Bhim, M. (2019). Does electoral authoritarianism persist? A comparison of recent
elections in Fiji, Seychelles, and Maldives. In: J. I. Lahai, K. von Strokirch,
H. Brasted, and H. Ware (Eds.), Governance and Political Adaptation in Fragile
States, pp. 243–270. Cham, Switzerland: Palgrave Macmillan.
Bieber, F. (2003). The Serbian opposition and civil society: Roots of the delayed
transition in Serbia. International Journal of Politics, Culture, and Society,
17(1), 73–90.
Birch, S. (2003). Electoral Systems and Political Transformation in Post-Communist
Europe. New York, NY: Palgrave Macmillan.
Birch, S. (2011). Electoral Malpractice. Oxford, UK: Oxford University Press.
Blais, A. and Carty R. K. (1987). The impact of electoral formulae on the creation
of majority governments. Electoral Studies, 6(3), 209–218.

6"×9"
Blais, A. and Massicotte, L. (2002). Electoral systems. In: L. Leduc, R. Niemi, and
P. Norris (Eds.), Comparing Democracies 2: New Challenges in the Study of
Elections and Voting, pp. 40–69. Thousand Oaks, CA: Sage.
Blaydes, L. (2010). Elections and Distributive Politics in Mubarak’s Egypt. Cambridge,
UK: Cambridge University Press.
Blondel, J. (1968). Party systems and patterns of government in Western democra-
cies. Canadian Journal of Political Science, 1(2), 180–203.
Blondel, J. and Thiebault, J.-L. (2020). Parliamentary and presidential systems:
The role of parties and the danger of authoritarian populism. In: I. Crewe and
D. Sanders (Eds.), Authoritarian Populism and Liberal Democracy, pp. 169–182.
Cham, Switzerland: Palgrave Macmillan.
Bochsler, D. (2010). Measuring party nationalisation: A new Gini-based indicator
that corrects for the number of units. Electoral Studies, 29(1), 155–168.
Boese, V. A. (2019). How (not) to measure democracy. International Area Studies
Review, 22(2), 95–127.
Bogaards, M. (2004). Counting parties and identifying dominant party systems in
Africa. European Journal of Political Research, 43(2), 173–197.
Bogaards, M. (2008). Dominant party systems and electoral volatility in Africa:
A comment on Mozaffar and Scarritt. Party Politics, 14(1), 113–130.
Bogaards, M. (2009). How to classify hybrid regimes? Defective democracy and
electoral authoritarianism. Democratization, 16(2), 399–423.
Bogaards, M. (2012). Where to draw the line? From degree to dichotomy in measures
of democracy. Democratization, 19(4), 690–712.
Boix, C., Miller, M., and Rosato, S. (2013). A complete data set of political regimes,
1800–2007. Comparative Political Studies, 46(12), 1523–1554.
Boix, C. and Svolik, M. W. (2013). The foundations of limited authoritarian govern-
ment: Institutions, commitment, and power-sharing in dictatorships. Journal
of Politics, 75(2), 300–316.
Bolt, J. and van Zanden, J. L. (2020). The Maddison project: Maddison style estimates
of the evolution of the world economy. A new 2020 update. Maddison-Project
Working Paper WP-15, October.
Bormann, N.-C. and Golder, M. (2013). Democratic electoral systems around the
world, 1946–2011. Electoral Studies, 32(2), 360–369.
Bozóki, A. and Ishiyama, J. T. (Eds.). (2002). Communist Successor Parties in Central
and Eastern Europe: Reform or Transmutation? Armonk, NY: M. E. Sharpe.
Brancati, D. (2008a). The origins and strength of regional parties. British Journal of
Political Science, 38(1), 135–159.

References 233
Brancati, D. (2008b). Winning alone: The electoral fate of independent candidates

worldwide. Journal of Politics, 70(3), 648–662.
Brancati, D. (2014). Democratic authoritarianism: Origins and effects. Annual
Review of Political Science, 17, 313–326.
Brinks, D. and Coppedge, M. (2006). Diffusion is no illusion: Neighbor emulation
in the third wave of democracy. Comparative Political Studies, 39(4), 463–489.
Brown, T. L. (1996). The Challenge to Democracy in Nepal: A Political History.
London, UK: Routledge.
Browne, E. C. and Hamm, K. E. (1996). Legislative politics and the paradox of
voting: Electoral reform in Fourth Republic. British Journal of Political Science,
26(2), 165–198.
Brownlee, J. (2007). Authoritarianism in an Age of Democratization. Cambridge,
Bryce, J. (2008) [1901]. Constitutions. Charleston, SC: BiblioBazaar.
Bueno de Mesquita, B., Smith, A., Siverson, R. M., and Morrow, J. D. (2003). The
Logic of Political Survival. Cambridge, MA: MIT Press.
Bulbeck, C. (1984). Socialism in the Seychelles. Australian Outlook, 38(1), 40–44.
Bunbongkarn, S. (1987). The Military in Thai Politics, 1981–86. Singapore: Institute
of Southeast Asian Studies.
Bunce, V. (1995). Should transitologists be grounded? Slavic Review, 54(1), 111–127.
Bunce, V. (1999). Subversive Institutions: The Design and the Destruction of Socialism
and the State. New York, NY: Cambridge University Press.
Bunce, V. and Wolchik, S. L. (2010). Defeating dictators: Electoral change and
stability in competitive authoritarian regimes. World Politics, 62(1), 43–86.
Calabrese, S. (2007). An explanation of the continuing federal government mandate
of single-member congressional districts. Public Choice, 130(1–2), 23–40.
Calingaert, D. (2006). Election rigging and how to fight it. Journal of Democracy,
17(3), 138–151.
Callinicos, A. (1987). Making History: Agency, Structure and Change in Social Theory.
Cambridge, UK: Polity Press.
Cameron, D. R. and Orenstein, M. A. (2012). Post-Soviet authoritarianism: The
influence of Russia in its ‘near abroad’. Post-Soviet Affairs, 28(1), 1–44.
Caramani, D. (2000). Elections in Western Europe Since 1815: Electoral Results by
Constituencies. London, UK: Macmillan Reference.
Caramani, D. (2004). The Nationalization of Politics: The Formation of National
Electorates and Party Systems in Western Europe. Cambridge, UK: Cambridge
University Press.

6"×9"
Carey, J. M. and Hix, S. (2011). The electoral sweet spot: Low-magnitude propor-
tional electoral systems. American Journal of Political Science, 55(2), 383–397.
Carothers, C. (2018). The surprising instability of competitive authoritarianism.
Journal of Democracy, 29(4), 129–135.
Casal Bértoa, F. (2017). Political parties or party systems? Assessing the ‘myth’ of
institutionalisation and democracy. West European Politics 40(2), 402–429.
Case, W. (2007). Semi-democracy and minimalist federalism in Malaysia. In:
B. He, B. Galligan, and T. Inoguchi (Eds.), Federalism in Asia, pp. 124–143.
Cheltenham, UK: Edward Elgar.
Cassani, A. and Tomini, L. (2019). Autocratization in Post-Cold War Political Regimes.
New York, NY: Palgrave Macmillan.
Cavatorta, F. (2009). ‘Divided they stand, divided they fail’: Opposition politics in
Morocco. Democratization, 16(1), 137–156.
Chafer, T. (2002). The End of Empire in French West Africa: France’s Successful
Decolonization. Oxford, UK: Berg.
Chambers, P. and Waitoolkiat, N. (2016). The resilience of monarchised military in
Thailand. Journal of Contemporary Asia, 46(3), 425–444.
Chand, V. K. (2001). Mexico’s Political Awakening. Notre Dame, IN: University of
Notre Dame Press.
Cheibub, J. A. and Chernykh, S. (2009). Are semi-presidential constitutions bad for
democratic performance? Constitutional Political Economy, 20(3), 202–229.
Cheibub, J. A. and Hays, J. C. (2017). Elections and civil war in Africa. Political
Science Research and Methods 5(1), 81–102.
Cheibub, J. A., Gandhi, J., and Vreeland, J. R. (2010). Democracy and dictatorship
revisited. Public Choice, 143(1), 67–101.
Chhibber, P. and Kollman, K. (1998). Party aggregation and the number of parties in
India and the United States. American Political Science Review, 92(2), 329–342.
Chhibber, P. and Kollman, K. (2004). The Formation of National Party Systems:
Federalism and Party Competition in Canada, Great Britain, India, and the
United States. Princeton, NJ: Princeton University Press.
Chhibber, P. and Murali, G. (2006). Duvergerian dynamics in the Indian states:
Federalism and the number of parties in the state assembly elections. Party
Politics, 12(1), 5–34.
Chhibber, P. and Nooruddin, I. (2004). Do party systems count? The number of
parties and government performance in the Indian states. Comparative Political
Studies, 37(2), 152–187.

References 235
Clark, J. F. (1998). Foreign intervention in the civil war of the Congo Republic. Issue:
A Journal of Opinion, 26(1), 31–36.
Clark, W. R. and Golder, M. (2006). Rehabilitating Duverger’s theory: Testing the
mechanical and strategic modifying effects of electoral laws. Comparative
Codato, A. N. (Ed.) (2006). Political Transition and Democratic Consolidation: Studies
on Contemporary Brazil. New York, NY: Nova Science Publishers.
Cohen, G. B. (2007). Nationalist politics and the dynamics of state and civil society in
the Habsburg Monarchy, 1867–1914. Central European History, 40(2), 241–278.
Collier, B. R. (1978). Parties, coups, and authoritarian rule: Patterns of political
change in Tropical Africa. Comparative Political Studies, 11(1), 62–93.
Collier, D. (2008). Typologies: Forming concepts and creating categorical variables.
In: J. M. Box-Steffensmeier, H. E. Brady, and D. Collier (Eds.), Oxford Handbook
of Political Methodology, pp. 152–173. Oxford, UK: Oxford University Press.
Colomer, J. M. (2007). On the origins of electoral systems and political parties: The
role of elections in multi-member districts. Electoral Studies, 26(2), 262–273.
Conaghan, C. M. and Espinal, R. (1990). Unlikely transitions to uncertain regimes?
Democracy without compromise in the Dominican Republic and Ecuador.
Journal of Latin American Studies, 22(3), 553–574.
Coppedge, M. (1998). The dynamic diversity of Latin American party systems.
Party Politics, 4(4), 547–568.
Coppedge, M., Gerring, J., Altman, D., Bernhard, M., Fish, M. S., Hicken, A.,
Kroenig, M., Lindberg, S. I., McMann, K., Paxton, P., Semetko, H. A., Skaaning,
S.-E., Staton, J., and Teorell, J. (2011). Conceptualizing and measuring democ-
racy: A new approach. Perspectives on Politics, 9(2), 247–267.
Coppedge, M., Gerring, J., Lindberg, S. I., Skaaning, S.-E., and Teorell, J. (2017).
V-Dem comparisons and contrasts with other measurement projects. University
of Gothenburg, Varieties of Democracy Institute: Working Paper No. 45.
Coppedge, M., Gerring, J., Knutsen, C. H., Lindberg, S. I., Teorell, J., Altman, D.,
Bernhard, M., Fish, M. S., Glynn, A., Hicken, A., Lührmann, A., Marquardt, K.
L., McMann, K., Paxton, P., Pemstein, D., Seim, B., Sigman, R., Skaaning, S-E.,
Staton, J., Wilson, S., Cornell, A., Alizada, N., Gastaldi, L., Gjerløw, H.,
Hindle, G., Ilchenko, N., Maxwell, L., Mechkova, V., Medzihorsky, J.,
von Römer, J., Sundström, A., Tzelgov, E., Wang, Y.-T., Wig, T., and Ziblatt, D.
(2020). V-Dem [Country–Year/Country–Date] Dataset v10. Varieties of
Democracy (V-Dem) Project. Available from http://doi.org/10.23696/vdemds20.

6"×9"
Cox, G. W. (1990). Centripetal and centrifugal incentives in electoral systems.

American Journal of Political Science 34(4), 903–935.
Cox, G. W. (1997). Making Votes Count: Strategic Coordination in the World’s Electoral
Systems. New York, NY: Cambridge University Press.
Cox, K. E. and Schoppa, L. J. (2002). Interaction effects in mixed-member electoral
systems: Theory and evidence from Germany, Japan, and Italy. Comparative
Croissant, A. and Kamerling, J. (2013). Why do military regimes institutionalize?
Constitution-making and elections as political survival strategy in Myanmar.
Asian Journal of Political Science, 21(2), 105–125.
Croissant, A., Kuehn, D., Chambers, P., and Wolf, S. O. (2010). Beyond the fallacy
of coup-ism: Conceptualizing civilian control of the military in emerging
democracies. Democratization, 17(5), 950–975.
Crouch, H. (1979). Patrimonialism and military rule in Indonesia. World Politics,
31(4), 571–587.
Crouch, M. (2019). The Constitution of Myanmar: A Contextual Analysis. Oxford,
UK: Hart Publishing.
Daalder, H. (2002). Parties: Denied, dismissed, or redundant? A critique. In:
R. Gunther, J. Linz, and J. R. Montero (Eds.), Political Parties: Old Concepts and
New Challenges, pp. 39–57. Oxford, UK: Oxford University Press.
Daddieh, C. K. (2001). Elections and ethnic violence in Côte d’Ivoire: The unfinished
business of succession and democratic transition. Journal of Eastern African
Studies, 29(1/2), 14–19.
Dahl, R. A. (1989). Democracy and Its Critics. New Haven, CT: Yale University Press.
Däubler, T. and Hix, S. (2018). Ballot structure, list flexibility and policy represen-
tation. Journal of European Public Policy, 25(12), 1798–1816.
De Jager, N. and du Toit, P. (2013). Introduction. In: N. De Jager and P. du Toit
(Eds.), Friend or Foe? Dominant Party Systems in Southern Africa: Insights
from the Developing World, pp. 3–20. Tokyo, Japan: United Nations University
Press.
Deas, M. (1996). The role of the church, the army and the police in Colombian
elections, c. 1850–1930. In: E. Posada-Carbó (Ed.), Elections before Democracy:
The History of Elections in Europe and Latin America, pp. 163–180. Basingstoke,
UK: Palgrave Macmillan.
Decalo, S. (1985a). Socio-economic constraints on radical action in the People’s
Republic of Congo. Journal of Communist Studies, 1(3–4), 39–57.

References 237
Decalo, S. (1985b). The morphology of radical military rule in Africa. Journal of

Communist Studies, 1(3–4), 122–144.
Dekmejian, R. H. (1968). The U. A. R. National Assembly — A pioneering experi-
ment. Middle Eastern Studies, 4(4), 361–337.
Del Sordi, A. (2016). Legitimation and the party of power in Kazakhstan. In:
M. Brusis, J. Ahrens, and M. S. Wessel (Eds.), Politics and Legitimacy in Post-
Soviet Eurasia, pp. 72–96. Basingstoke, UK: Palgrave Macmillan.
Deltas, G. (2003). The small-sample bias of the Gini coefficient: Results and implica-
tions for empirical research. Review of Economics and Statistics, 85(1), 226–234.
Demchenko, O. and Golosov, G. V. (2016). Federalism, gubernatorial power and
the incorporation of subnational authoritarianism in Russia: A theory-testing
empirical inquiry. Acta Politica, 51(1), 61–79.
Despres, L. A. (Ed.). (1975). Ethnicity and Resource Competition in Plural Societies.
The Hague, Netherlands: Mouton & Co.
Detalle, R. and Hiltermann, J. (1993). The Yemeni elections up close. Middle East
Report. No. 185, 8–12.
Diamond, L. (1999). Developing Democracy: Toward Consolidation. Baltimore, MD:
Johns Hopkins University Press.
Diaz-Cayeros, A., Magaloni, B. and Weingast, B. R. (2003). Tragic brilliance:
Equilibrium party hegemony in Mexico. Available from http://dx.doi.
org/10.2139/ssrn.1153510.
Dickson, B. J. (1993). The lessons of defeat: The reorganization of the Kuomintang
on Taiwan, 1950–52. The China Quarterly, 133, 56–84.
DiMaggio, P. J. and Powell, W. W. (1983). The iron cage revisited: Institutional
isomorphism and collective rationality in organizational fields. American
Sociological Review, 48(2), 147–160.
Dinas, E. and Northmore-Ball, K. (2020). The ideological shadow of authoritarian-
ism. Comparative Political Studies, 53(12), 1957–1991.
Donno, D. (2013). Elections and democratization in authoritarian regimes. American
Journal of Political Science, 57(3), 703–716.
Downs, A. (1957). An Economic Theory of Democracy. New York, NY: Harper and Row.
Drummond, A. J. (2006). Electoral volatility and party decline in western democ-
racies: 1970–1995. Political Studies, 54(3), 628–647.
Dudoignon, S. (1997). Political parties and forces in Tajikistan, 1989–1993.
In: M.-R. Djalili, F. Grare, and S. Akiner (Eds.), Tajikistan: The Trails of
Independence, pp. 52–87. New York, NY: St. Martin’s Press.

6"×9"
Dumont, P. and Caulier, J.-F. (2003). The ‘effective number of relevant parties’:
How voting power improves Laakso-Taagepera’s index. Center for Research in
Economics (CEREC), Facultés Universitaires Saint-Louis, Brussels, Working
Paper 2003/7.
Dunleavy, P. and Boucek, F. (2003). Constructing the number of parties. Party
Politics, 9(3), 291–315.
Durac, V. (2011). The joint meeting parties and the politics of opposition in Yemen.
British Journal of Middle Eastern Studies, 38(3), 343–365.
Duverger, M. (1954). Political Parties: Their Organisation and Activity in the Modern
State. London: Methuen.
Duverger, M. (1980). A new political system model: Semi-presidential government.
European Journal of Political Research, 8(2), 165–187.
Eggers, A. C. and Fouirnaies, A. B. (2014). Representation and district magnitude
in plurality systems. Electoral Studies, 33(1), 267–277.
Eklöf, S. (1997). The 1997 general election in Indonesia. Asian Survey, 37(12),
1181–1196.
Elazar, D. J. (1991). Federal Systems of the World: A Handbook of Federal, Confederal
and Autonomy Arrangements. London: Longman.
Elgie, R. (1999). The politics of semi-presidentialism. In: Elgie, R. (Ed.), Semi-
Presidentialism in Europe, pp. 1–21. Oxford: Oxford University Press.
Elgie, R. (2005). From Linz to Tsebelis: Three waves of presidential/parliamentary
studies? Democratization, 12(1), 106–122.
Elgie, R. (2008). The perils of semi-presidentialism. Are they exaggerated?
Democratization, 15(1), 49–66.
Elgie, R. (2011). Semi-Presidentialism: Sub-Types and Democratic Performance.
Oxford: Oxford University Press.
Elgie, R. (n. d.). The Semi-Presidential One. Retrieved March 9, 2021, from http://
www.semipresidentialism.com.
Elklit, J. and Reynolds, A. (2005). A framework for the systematic study of election
quality. Democratization, 12(2), 147–162.
Epkenhans, T. (2018). The Islamic Renaissance Party of Tajikistan: Episodes of
Islamic activism, postconflict accommodation, and political marginalization.
In: Laruelle, M. (Ed.), Being Muslim in Central Asia, pp. 85–109. Leiden,
Netherlands: Brill.
Epstein, L. D. (1980). Political Parties in Western Democracies. New Brunswick, NJ:
Transaction Books.

References 239
Ersson, S., Janda, K. and Lane, J. E. (1985). Ecology of party strength in Western
Europe: A regional analysis. Comparative Political Studies, 18(2), 170–205.
Etges, A. (2011). All that glitters is not gold: The 1953 coup against Mohammed
Mossadegh in Iran. Intelligence and National Security, 26(4), 495–508.
Fainsod, M. (1958). Smolensk Under Soviet Rule. Cambridge, MA: Harvard
University Press.
Falleti, T. G. (2011). Varieties of authoritarianism: The organization of the military
state and its effects on federalism in Argentina and Brazil. Studies in Comparative
International Development, 46(2), 137–162.
Faridul Haq, M. (1968). The constitution of Iran in its historical setting. Pakistan
Horizon, 21(1), 3–13.
Farrell, D. M. (2001). Electoral Systems: A Comparative Introduction. 2nd Edition.
New York, NY: Palgrave.
Fearon, J. (2003). Ethnic and cultural diversity by country. Journal of Economic
Growth, 8(2), 195–222.
Fernandes, J. M. and Magalhães, P. C. (2016). Government survival in semi-
presidential regimes. European Journal of Political Research, 55(1), 61–80.
Ferrara, F., Herron, E. S. and Nishikawa, M. (2005). Mixed Electoral Systems:
Contamination and Its Consequences. New York, NY: Palgrave Macmillan.
Finer, S. E. (1983). Perspectives in the world history of government — A prolegom-
enon. Government and Opposition, 18(1), 3–22.
Fishman, R. M. (1990). Rethinking state and regime: Southern Europe’s transition
to democracy. World Politics, 42(3), 422–440.
Fitch, J. S. (1998). The Armed Forces and Democracy in Latin America. Baltimore,
MD: Johns Hopkins University Press.
Fortin-Rittberger, J. (2014). The role of infrastructural and coercive state capacity
in explaining different types of electoral fraud. Democratization, 21(1), 95–117.
Frantz, E. and Kendall-Taylor, A. (2014). A dictator’s toolkit: Understanding how
co-optation affects repression in autocracies. Journal of Peace Research, 51(3),
332–346.
Frantz, E. and Kendall-Taylor, A. (2017). Pathways to democratization in personalist
dictatorships. Democratization, 24(1), 20–40.
Franzmann, S. T. (2011). Competition, contest, and cooperation: The analytic
framework of the issue market. Journal of Theoretical Politics, 23(3), 317–343.
Friedrich, C. J. and Brzezinski, Z. K. (1961). Totalitarian Dictatorship and Autocracy.
2nd Edition. New York, NY: Praeger.

6"×9"
Frye, T., Gehlbach, S., Marquardt, K. L., and Reuter, O. J. (2017). Is Putin’s popularity
real? Post-Soviet Affairs, 33(1), 1–15.
Fukui, H. (1988). Electoral laws and the Japanese party system. In: G. L. Bernstein
and H. Fukui (Eds.), Japan and the World: Essays in Japanese History and Politics
in Honour of Ishida Takeshi, pp. 119–143. Basingstoke: Macmillan, 1988.
Gallagher, M. (1991). Proportionality, disproportionality and electoral systems.
Electoral Studies, 10(3), 33–51.
Gallagher, M. (1992). Comparing proportional representation electoral systems:
Quotas, thresholds, paradoxes and majorities. British Journal of Political
Science, 22(4), 469–496.
Gandhi, J. (2008). Political Institutions under Dictatorship. Cambridge: Cambridge
University Press.
Gandhi, J. and Lust-Okar, E. (2009). Elections under authoritarianism. Annual
Gandhi, J. and Ong, E. (2019). Committed or conditional democrats? Opposition
dynamics in electoral autocracies. American Journal of Political Science, 63(4),
948–963.
Gandhi, J. and Przeworski, A. (2006). Cooperation, cooptation, and rebellion under
dictatorships. Economics & Politics, 18(1), 1–26.
Gandhi, J. and Przeworski, A. (2007). Authoritarian institutions and the survival of
autocrats. Comparative Political Studies, 40(11), 1279–1301.
Gandhi, J. and Reuter, O. J. (2013). The incentives for pre-electoral coalitions in
non-democratic elections. Democratization, 20(1), 137–159.
Geddes, B. (1999). What do we know about democratization after 20 years? Annual
Geddes, B. (2003). Paradigms and Sand Castles: Theory Building and Research
Design in Comparative Politics. Ann Arbor, MI: University of Michigan
Press.
Geddes, B., Wright, J., and Frantz, E. (2014). Autocratic breakdown and regime
transitions: A new data set. Perspectives on Politics, 12(2), 313–331.
Geddes, B., Wright, J., and Frantz, E. (2018). How Dictatorships Work: Power,
Personalization, and Collapse. Cambridge: Cambridge University Press.
Gehlbach, S., Sonin, K., and Svolik, M. W. (2016). Formal models of nondemocratic
politics. Annual Review of Political Science, 19, 565–584.
Gel’man, V. (2005). Political opposition in Russia: A dying species? Post-Soviet
Affairs, 21(3), 226–246.

References 241
Ghanem, A. and Mustafa, M. (2010). Strategies of electoral participation by Islamic

movements: The Muslim Brotherhood and parliamentary elections in Egypt
and Jordan, November 2010. Contemporary Politics, 17(4), 393–409.
Gibson, E. L. and Suarez-Cao, J. (2010). Federalized party systems and subna-
tional party competition: Theory and an empirical application to Argentina.
Comparative Politics, 43(1), 21–39.
Gilbert, L. and Mohseni, P. (2011). Beyond authoritarianism: The conceptualization
of hybrid regimes. Studies in Comparative International Development, 46(3),
270–297.
Gilley, B. Democratic enclaves in authoritarian regimes. Democratization, 17(3),
389–415.
Ginsburg, T. and Simpser, A. (Eds.). (2013). Constitutions in Authoritarian Regimes.
Cambridge: Cambridge University Press.
Gleditsch, K. S. and Ward, M. D. (1997). Double take: A re-examination of democ-
racy and autocracy in modern polities. Journal of Conflict Resolution, 41(3),
361–383.
Golosov, G. V. (1996). Modes of communist rule, democratic transition, and party
system formation in four East European countries. Donald W. Treadgold Paper
in Russian, East European and Central Asian Studies No. 9. Seattle, WA, Henry
M. Jackson School of International Studies, University of Washington.
Golosov, G. V. (1998). Who survives? Party origins, organisational development,
and electoral performance in post-Communist Russia. Political Studies, 46(3),
511–543.
Golosov, G. V. (2002). Party support or personal resources? Factors of success in the
plurality portion of the 1999 national legislative elections in Russia. Communist
and Post-Communist Studies, 35(1), 23–38.
Golosov, G. V. (2003a). Electoral systems and party formation in Russia: A cross-
regional analysis. Comparative Political Studies, 36(8), 912–935.
Golosov, G. V. (2003b). The vicious circle of party underdevelopment in Russia:
The regional connection. International Political Science Review, 24(4),
427–444.
Golosov, G. V. (2006a). Disproportionality by proportional design: Seats and votes
in Russia’s regional legislative elections. Europe-Asia Studies, 58(1), 25–55.
Golosov, G. V. (2006b). The structure of party alternatives and voter choice in
Russia: Evidence from the 2003–2004 regional legislative elections. Party
Politics, 12(6), 707–725.

6"×9"
Golosov, G. V. (2010). The effective number of parties: A new approach. Party

Politics, 16(2), 171–192.
Golosov, G. V. (2011a). Party system classification: A methodological inquiry. Party
Politics, 17(5), 539–560.
Golosov, G. V. (2011b). The regional roots of electoral authoritarianism in Russia.
Europe-Asia Studies, 63(4), 623–639.
Golosov, G. V. (2013a). Authoritarian party systems: Patterns of emergence,
sustainability and survival. Comparative Sociology, 12(5), 617–644.
Golosov, G. V. (2013b). Proportional representation and authoritarianism: Evidence
from Russia’s regional election law reform. Representation, 49(1), 83–95.
Golosov, G. V. (2013c). Towards a classification of the world’s democratic party
systems, Step 1: Identifying the units. Party Politics, 19(1), 122–142.
Golosov, G. V. (2014a). Authoritarian electoral engineering and its limits: A curious
case of the Imperiali highest averages method in Russia. Europe-Asia Studies,
66(10), 1611–1628.
Golosov, G. V. (2014b). Co-optation in the process of dominant party system build-
ing: The case of Russia. East European Politics, 30(2), 271–285.
Golosov, G. V. (2014c). The last resort of political regionalism: Electoral blocs in
Russia’s regional legislative elections, 2003–2005. East European Politics, 30(1),
71–85.
Golosov, G. V. (2014d). The territorial genealogies of Russia’s political parties
and the transferability of political machines. Post-Soviet Affairs, 30(6),
464–480.
Golosov, G. V. (2014e). Towards a classification of the world’s democratic party
systems, Step 2: Placing the units into categories. Party Politics, 20(1), 3–15.
Golosov, G. V. (2015a). Do spoilers make a difference? Instrumental manipulation
of political parties in an electoral authoritarian regime, the case of Russia. East
European Politics, 31(2), 170–186.
Golosov, G. V. (2015b). Factors of party system fragmentation: A cross-national
study. Australian Journal of Political Science, 50(1), 42–60.
Golosov, G. V. (2015c). The idiosyncratic dynamics of party system nationalization
in Russia. Post-Soviet Affairs, 31(5), 397–419.
Golosov, G. V. (2016a). Factors of party system nationalization. International Political
Science Review, 37(2), 246–260.
Golosov, G. V. (2016b). Party system nationalization: The problems of measurement
with an application to federal states. Party Politics, 22(3), 278–288.

References 243
Golosov, G. V. (2016c). Voter volatility in electoral authoritarian regimes: Testing

the ‘tragic brilliance’ thesis. Comparative Sociology, 15(5), 535–559.
Golosov, G. V. (2016d). Why and how electoral systems matter in autocracies.
Australian Journal of Political Science, 51(3), 367–385.
Golosov, G. V. (2017). Party systems, electoral systems, and legislative fragmentation:
A cross-national empirical study. Party Politics, 23(5), 487–497.
Golosov, G. V. (2018). Party nationalization and the translation of votes into seats
under single-member plurality electoral rules. Party Politics, 24(2), 118–128.
Golosov, G. V. (2020). The five shades of grey: Party systems and authoritarian insti-
tutions in post-Soviet Central Asian states. Central Asian Survey, 39(3), 285–302.
Golosov, G. V., Gushchina, K., and Kononenko, P. (2016). Russia’s local government
in the process of authoritarian regime transformation: Incentives for the survival
of local democracy. Local Government Studies, 42(4), 507–526.
Golosov, G. V. and Kalinin, K. (2017). Presidentialism and legislative fragmentation:
Beyond coattail effects. British Journal of Politics and International Relations,
19(1), 113–133.
Golosov, G. V. and Konstantinova, M. (2016). Gubernatorial powers in Russia: The
transformation of regional institutions under the centralizing control of the
federal authorities. Problems of Post-Communism, 63(4), 241–252.
Golosov, G. V. and Ponarin, E. (1999). Regional bases of party politics: A measure
and its implications for the study of party system consolidation in new democ-
racies. Robert Schuman Center for Advanced Studies Working Paper, 99/25.
Badia Fiesolana, European University Institute.
Golosov, G. V. and Turchenko, M. (2021). Countering the “sweep effect”: Opposition
voter coordination versus electoral system effects in authoritarian local elec-
tions. Communist and Post-Communist Studies, 54(1–2), 66–82.
Goodwin-Gill, G. S. (2006). Free and Fair Elections. Geneva, Switzerland: Inter-
Parliamentary Union.
Greene, K. E. (2007). Why Dominant Parties Lose: Mexico’s Democratization in
Comparative Perspective. New York, NY: Cambridge University Press.
Gregory, M. (1981). Zimbabwe 1980: Politicisation through armed struggle and
electoral mobilization. Journal of Commonwealth & Comparative Politics,
19(1), 63–94.
Grofman, B. (1989). The comparative analysis of coalition formation and duration:
Distinguishing between-country and within-country effects. British Journal of

6"×9"
Grofman, B. (2006). The impact of electoral laws on political parties. In: D. Wittman
and B. Weingast (Eds.), Oxford Handbook of Political Economy, pp. 102–118.
Oxford: Oxford University Press.
Grofman, B., Chiaramonte, A., D’Alimonte, R., and Feld, S. L. (2004). Comparing and
contrasting the uses of two graphical tools for displaying patterns of m ultiparty
competition: Nagayama diagrams and simplex representations. Party Politics,
10(3), 273–299.
Gros, J.-G. (1995). The hard lessons of Cameroon. Journal of Democracy, 6(3),
112–127.
Gross, D. A. and Sigelman, L. (1984). Comparing party systems: A multidimensional
approach. Comparative Politics, 16(4), 463–479.
Grzymala-Busse, A. (2002). Redeeming the Communist Past: The Regeneration of
Communist Parties in East Central Europe. New York, NY: Cambridge University
Press.
Grzymala-Busse, A. (2019). Hoist on their own petards? The reinvention and collapse
of authoritarian successor parties. Party Politics, 25(4), 569–582.
Grzymala-Busse, A. (2020). Consequences of authoritarian party exit and rein-
vention for democratic competition. Comparative Political Studies, 53(10–11),
1704–1737.
Gunther, R. and Diamond, L. (2003). Species of political parties: A new typology.
Hadenius, A. and Teorell, J. (2007). Pathways from authoritarianism. Journal of
Democracy, 18(1), 143–156.
Hagopian, F. (1996). Traditional Politics and Regime Change in Brazil. Cambridge:
Cambridge University Press.
Handy, J. (1984). Gift of the Devil: A History of Guatemala. Toronto, Canada: Between
the Lines Press.
Harbers, I. (2010). Decentralization and the development of nationalized party
systems in new democracies: Evidence from Latin America. Comparative
Harding, A. and Leyland, P. (2011). The Constitutional System of Thailand:
A Contextual Analysis. London, UK: Hart Publishing.
Hartlyn, J. (1988). The Politics of Coalition Rule in Colombia. New York, NY:
Hatchard, J. (1993). Re-establishing a multi-party state: Some constitutional lessons
from the Seychelles. Journal of Modern African Studies, 31(4), 601–612.

References 245
Hazis, F. S. (2018). Domination, contestation, and accommodation: 54 years of Sabah

and Sarawak in Malaysia. Southeast Asian Studies, 7(3), 341–361.
Hearl, D., Budge, I., and Pearson, B. (1996). Distinctiveness of regional voting:
A comparative analysis across the European Community (1979–1993). Electoral
Studies, 15(2), 167–182.
Helle, S.-E. and Rakner, L. (2014). Grabbing an election: Abuse of state resources in
the 2011 elections in Uganda. In: T. Søreide and A. Williams (Eds.), Corruption,
Grabbing and Development: Real World Challenges, pp. 161–172. Cheltenham,
UK: Edward Elgar.
Hellmeier, S. and Weidmann, N. B. (2020). Pulling the strings? The strategic use of
pro-government mobilization in authoritarian regimes. Comparative Political
Studies, 53(1), 71–108.
Heo, U. and Stockton, H. (2005). The impact of democratic transition on elections
and parties in South Korea. Party Politics, 11(6), 674–688.
Hermet, G., Rose, R., and Rouquie, A. (Eds.) (1978). Elections without Choice.
New York, NY: John Wiley.
Herron, E. S. and Nishikawa, M. (2001). Contamination effects and the number of
parties in mixed-superposition electoral systems. Electoral Studies, 20(1), 63–86.
Hewison, K. (1997). The monarchy and democratization. In: K. Hewison (Ed.),
Political Change in Thailand: Democracy and Participation, pp. 58–74. London,
UK: Routledge.
Hicken, A. and Kuhonta, E. M. (2011). Shadow from the past: Party system institu-
tionalization in Asia. Comparative Political Studies, 44(5), 572–597.
Hicken, A. and Stoll, H. (2008). Electoral rules and the size of the prize: How
political institutions shape presidential party systems. Journal of Politics, 70(4),
1109–1127.
Hicken, A. and Stoll, H. (2011). Presidents and parties: How presidential elections
shape coordination in legislative elections. Comparative Political Studies, 44(7),
854–883.
Hicken, A. and Stoll, H. (2013). Are all presidents created equal? Presidential powers
and the shadow of presidential elections. Comparative Political Studies, 46(3),
291–319.
Higashijima, M. and Kasuya, Y. (2016). The peril of parliamentarism? Executive-
legislative relations and the transition to democracy from electoral authoritar-
ian rule. EUI Working Paper MWP 2016/01, Max Weber Programme. Badia
Fiesolana, European University Institute.

6"×9"
Higley, J. and Gunther, R. (Eds.) (1992). Elites and Democratic Consolidation in Latin
America and Southern Europe. Cambridge, UK: Cambridge University Press.
Hilgers, T. (2008). Causes and consequences of political clientelism: Mexico’s PRD
in comparative perspective. Latin American Politics and Society, 50(4), 123–153.
Hill, E., Sobolewska, M., Wilks-Heeg, S., and Borkowska, M. (2017). Explaining
electoral fraud in an advanced democracy: Fraud vulnerabilities, opportunities
and facilitating mechanisms in British elections. British Journal of Politics and
International Relations, 19(4), 772–789.
Hosking, G. A. (1973). The Russian Constitutional Experiment: Government and
Duma, 1907–1914. New York, NY: Cambridge University Press.
Howard, M. M. and Roessler, P. G. (2006). Liberalizing electoral outcomes in compet-
itive authoritarian regimes. American Journal of Political Science, 50(2), 365–381.
Hoyt, K. (2004). Parties and pacts in contemporary Nicaragua. In: D. Close and
K. Deonandan (Eds.), Undoing Democracy: The Politics of Electoral Caudillismo,
pp. 17–42. Lanham, MD, Lexington Books.
Hsieh, J. F. and Niemi, R. G. (1999). Can Duverger’s law be extended to SNTV? The
case of Taiwan’s Legislative Yuan elections. Electoral Studies, 18(1), 101–116.
Huntington, S. P. (1968). Political Order in Changing Societies. New Haven, CT:
Yale University Press.
Huntington, S. P. (1993). The Third Wave: Democratization in the Late Twentieth
Century. Norman, OK: University of Oklahoma Press.
Hutchcroft, P. D. and Rocamora, J. (2003). Strong demands and weak institutions:
The origins and evolution of the democratic deficit in the Philippines. Journal
of East Asian Studies, 3(2), 259–292.
Hutt, M. (2006). Nepal and Bhutan in 2005: Monarchy and democracy, can they
co-exist? Asian Survey, 46(1), 120–124.
Ihalainen, P., Ilie, C., and Palonen, K. (Eds.) (2016). Parliament and Parliamentarism:
A Comparative History of a European Concept. New York, NY: Berghahn.
Isaacs, R. and Whitmore, S. (2014). The limited agency and life-cycles of personalized
dominant parties in the post-Soviet space: The cases of United Russia and Nur
Otan. Democratization, 21(4), 699–721.
Ishiyama, J. T. (1999). The communist successor parties and party organizational
development in post-communist politics. Political Research Quarterly, 51(1),
87–112.
Ishiyama, J. T. (2005). The former Marxist Leninist parties in Africa after the end
of the Cold War. Acta Politica, 40(4), 459–479.

References 247
Iskandaryan, A. (2018). The velvet revolution in Armenia: How to lose power in

two weeks. Demokratizatsiya: The Journal of Post-Soviet Democratization,
26(4), 465–482.
Janda, K. (1980). Political Parties: A Cross-National Survey. New York, NY: Free
Press.
Janos, A. (1970). The one-party state and social mobilization: East Europe between
the wars. In: S. P. Huntington and C. Moore (Eds.), Authoritarian Politics in
Modern Society: The Dynamics of Established One-Party Systems, pp. 204–236.
New York, NY: Basic Books.
Jastramskis, M. (2019). Effects of the mixed parallel electoral system in Lithuania:
The worst of all worlds? Parliamentary Affairs, 72(3), 561–587.
Jenkins, D. (1984). Suharto and His Generals: Indonesian Military Politics, 1975–1983.
Ithaca, NY: Cornell University Press.
Jensen, N. M., Malesky, E., and Weymouth, S. (2014). Unbundling the relationship
between authoritarian legislatures and political risk. British Journal of Political
Science, 44(3), 655–684.
Joffe, G. (1988). Morocco: Monarchy, legitimacy and succession source. Third World
Quarterly, 10(1), 201–228.
Johnson, K. F. (1971). The 1966 and 1970 elections in Guatemala: A comparative
analysis. World Affairs, 134(1), 34–50.
Jones, M. P. (1994). Presidential election laws and multipartism in Latin America.
Political Research Quarterly, 47(1), 41–57.
Jones, M. P. and Mainwaring, S. (2003). The nationalization of parties and party
systems: An empirical measure and an application to the Americas. Party
Politics, 9(2), 139–166.
Joseph, R. (1997). Democratization in Africa after 1989: Comparative and theoretical
perspectives. Comparative Politics, 29(3), 363–382.
Kahn, J. (2002). Federalism, Democratization, and the Rule of Law in Russia. Oxford,
UK: Oxford University Press.
Kailitz, S. (2013). Classifying political regimes revisited: Legitimation and durability.
Karl, T. L. (1989). The Christian Democratic Party and the prospects for democrati-
zation in El Salvador. In: J. L. Flora and E. Torres-Rivas (Eds.), Central America:
Sociology of Developing Societies, pp. 140–164. London, UK: Palgrave.
Karl, T. L. (1990). Dilemmas of democratization in Latin America. Comparative
Politics, 23(1), 1–21.

6"×9"
Karl, T. L. and Schmitter, P. (1991). Modes of transition in Latin America,

Southern and Eastern Europe. International Social Science Journal, 43(2),
269–284.
Kasenally, R. (2011). Mauritius: Paradise reconsidered. Journal of Democracy, 22(2),
160–169.
Kassem, M. (2004). Egyptian Politics: The Dynamics of Authoritarian Rule. Boulder,
CO: Lynne Rienner Publishers.
Kasuya, Y. and Moenius, J. (2008). The nationalization of party systems: Conceptual
issues and alternative district-focused measures. Electoral Studies, 27(1),
126–135.
Kasuya, Y. and Mori, K. (2019). Better regime cutoffs for continuous democracy
measures. University of Gothenburg, Varieties of Democracy Institute Working
Paper Series 2019: 25.
Kasuya, Y. and Sawasdee, S. N. (2019). The transformation of dominant parties in
Asia: Introduction to the special issue. Asian Journal of Comparative Politics,
4(1), 3–7.
Katz, R. S. (1980). A Theory of Parties and Electoral Systems. Baltimore, MD: Johns
Hopkins University Press.
Katz, R. S. and Mair, P. (1995). Changing models of party organization and party
democracy: The emergence of the cartel party. Party Politics, 1(1), 5–28.
Kaufmann, D. and Kraay, A. (n. d.). The worldwide governance indicators (WGI) proj-
ect. Retrieved March 20, 2021, from http://info.worldbank.org/governance/wgi.
Kefale, A. (2011). The (un)making of opposition coalitions and the challenge of
democratization in Ethiopia, 1991–2011. Journal of Eastern African Studies,
5(4), 681–701.
Khadka, N. (1986). Crisis in Nepal’s partyless Panchayat system: The case for more
democracy. Pacific Affairs, 59(3), 429–454.
Khadka, N. (1993). Democracy and development in Nepal: Prospects and challenges.
Pacific Affairs, 66(1), 44–71.
Kim, J. O. and Ohn M. G. (1992). A theory of minor-party persistence: Election
rules, social cleavage, and the number of political parties. Social Forces, 70(3),
575–599.
Kincaid, H. (1990). Defending laws in the social sciences. Philosophy of the Social
Sciences, 20(1), 56–83.
Kinzo, M. D. G. (1988). Legal Opposition Politics under Authoritarian Rule in Brazil:
The Case of the MDB 1966–79. Basingstoke, UK: Macmillan.

References 249
Kitschelt, H. (1995). Formation of party cleavages in post-communist democracies.

Kitschelt, H., Mansfeldova, Z., Markowski, R., and Tóka, G. (1999). Post-Communist
Party Systems: Competition, Representation, and Inter-Party Cooperation.
Kitschelt, H. and Wilkinson, S. (Eds.). (2007). Patrons, Clients, and Policies:
Patterns of Democratic Accountability and Political Competition. New York,
NY: Cambridge University Press.
Knight, A. (2005). Caciquismo in twentieth-century Mexico. In: A. Knight and W.
Pansters (Eds.), Caciquismo in Twentieth-Century Mexico, pp. 3–48. London,
UK: Institute for the Study of the Americas.
Kollman, K., Hicken, A., Caramani, D., Backer, D., and Lublin, D. (2019).
Constituency-Level Elections Archive [Data File and Codebook]. Ann Arbor,
MI: Center for Political Studies, University of Michigan [producer and distrib-
utor]. Retrieved from http://www.electiondataarchive.org.
Kryzanek, M. J. (1977). Diversion, subversion and repression: The strategies of
anti-opposition politics in Balaguer’s Dominican Republic. Caribbean Studies,
17(1–2), 83–103.
Kuenzi, M. and Lambright, G. (2001). Party system institutionalization in 30 African
countries. Party Politics, 7(4), 437–468.
Kynev, A. (2017). How the electoral policy of the Russian state predetermined the
results of the 2016 State Duma elections. Russian Politics, 2(2), 206–226.
Laakso, M. and Taagepera, R. (1979). ‘Effective’ number of parties: A measure with
application to West Europe. Comparative Political Studies, 12(1), 3–27.
Lachat, R. (2008). The impact of party polarization on ideological voting. Electoral
Studies, 27(4), 687–698.
Lambert, P. (2006). Nationalism and political hegemony in Paraguay. Bicentenario,
5(1), 45–62.
Lancaster, C. (2014). The iron law of Erdogan: The decay from intra-party democracy
to personalistic rule. Third World Quarterly, 35(9), 1672–1690.
Lane, J.-E. and Ersson, S. (2007). Party system instability in Europe: Persistent
differences in volatility between West and East? Democratization, 4(1), 92–110.
Laothamatas, A. (1998). Business and politics in Thailand: New patterns of influence.
Asian Survey, 28(4), 451–470.
LaPorte, J. (2015). Hidden in plain sight: Political opposition and hegemonic author-
itarianism in Azerbaijan. Post-Soviet Affairs, 31(4), 339–366.

6"×9"
Lawson, K. (1976). The Comparative Study of Political Parties. New York, NY:
St. Martin’s Press.
Lee, E. W. Y. (1999). Governing post-colonial Hong Kong: Institutional incongruity,
governance crisis, and authoritarianism. Asian Survey, 39(6), 940–959.
Lee, T. (2009). The armed forces and transitions from authoritarian rule: Explaining
the role of the military in 1986 Philippines and 1998 Indonesia. Comparative
Lemière, S. (2018). The downfall of Malaysia’s ruling party. Journal of Democracy,
29(4), 114–128.
Lemon, A. (1978). Electoral machinery and voting patterns in Rhodesia, 1962–1977.
African Affairs, 77(309), 511–530.
Levine, S. and Roberts, N. S. (2005). The constitutional structures and electoral
systems of Pacific island states. Commonwealth & Comparative Politics, 43(3),
276–295.
Lijphart, A. (1971). Comparative politics and the comparative method. American
Political Science Review, 65(3), 682–693.
Lijphart, A. (1984). Democracies: Patterns of Majoritarian and Consensus Government
in Twenty-One Countries. New Haven, CT: Yale University Press.
Lijphart, A. (1990). The political consequences of electoral laws, 1945–85. American
Lijphart, A. (1994). Electoral Systems and Party Systems: A Study of Twenty-Seven
Democracies, 1945–1990. Oxford, UK: Oxford University Press.
Lijphart, A. (1999). Patterns of Democracy: Government Forms and Performance in
Thirty-Six Countries. New Haven, CT: Yale University Press.
Likhtenchtein, A. and Yargomskaya, N. (2005). Duverger’s equilibrium under
limited competition: Russia’s parliamentary elections. Europe-Asia Studies,
57(8), 1169–1188.
Lindberg, S. I. (2007). Institutionalization of party systems? Stability and fluidity
among legislative parties in Africa’s democracies. Government and Opposition,
42(2), 215–241.
Lindberg, S. I. (2009). Democratization by elections? A mixed record. Journal of
Democracy, 20(3), 86–92.
Linhart, E., Raabe, J., and Statsch, P. (2019). Mixed-member proportional electoral
systems — The best of both worlds? Journal of Elections, Public Opinion and
Parties, 29(1), 21–40.
Linz, J. J. (1990). The perils of presidentialism. Journal of Democracy, 1(1), 51–69.

References 251
Lipset, S. M. and Rokkan, S. (1967). Cleavage structures, party systems, and voter align-
ments: An introduction. In: S. M. Lipset and S. Rokkan (Eds.), Party Systems and
Voter Alignments: Cross-National Perspectives, pp. 1–64. New York, NY: Free Press.
Loaeza, S. (2000). Uncertainty in Mexico’s protracted transition: The National Action
Party and its aversion to risk. Democratization, 7(3), 93–116.
Loveman, B. (1994). ‘Protected democracies’ and military guardianship: Political
transitions in Latin America, 1978–1993. Journal of Interamerican Studies and
World Affairs, 36(2), 105–189.
Loxton, J. and Mainwaring, S. (Eds.). (2018). Life after Dictatorship: Authoritarian
Successor Parties Worldwide. New York, NY: Cambridge University Press.
Lublin, D. (2017a). Electoral systems, ethnic diversity and party systems in devel-
oping democracies. Electoral Studies, 47, 84–93.
Lublin, D. (2017b). Electoral systems, ethnic heterogeneity and party system frag-
mentation. British Journal of Political Science, 47(2), 373–389.
Lührmann, A. and Lindberg, S. I. (2019). A third wave of autocratization is here:
What is new about it? Democratization, 26(7), 1095–1113.
Lührmann, A., Mechkova, V., Dahlum, S., Maxwell, L., Olin, M., Petrarca, C. S.,
Sigman, R., Wilson, M. C., and Lindberg, S. I. (2018). State of the world 2017:
Autocratization and exclusion? Democratization, 25(8), 1321–1340.
Lührmann, A., Tannenberg, M., and Lindberg, S. I. (2018). Regimes of the world
(RoW): Opening new avenues for the comparative study of political regimes.
Politics and Governance, 6(1), 60–77.
Luke, D. F. (1988). Continuity in Sierra Leone: From Stevens to Momoh. Third
World Quarterly, 10(1), 67–78.
Lundberg, T. C. (2009). Post-communism and the abandonment of mixed-member
electoral systems. Representation, 45(1), 15–27.
Lust-Okar, E. (2004). Divided they rule: The management and manipulation of
political opposition. Comparative Politics, 36(2), 159–179.
Lust-Okar, E. (2006). Elections under authoritarianism: Preliminary lessons from
Jordan. Democratization, 13(3), 456–471.
Lust-Okar, E. (2009). Reinforcing informal institutions through authoritarian
elections: Insights from Jordan. Middle East Law and Governance, 1(1), 3–37.
Lynch, D. (2001). The Tajik civil war and peace process. Civil Wars, 4(4), 49–72.
MacIntyre, A. (1971). Is a science of comparative politics possible? In: A. MacIntyre
(Ed.), Against the Self-Images of the Age: Essays in Ideology and Philosophy,
pp. 260–279. London, UK: Duckworth.

6"×9"
Maerz, S. F., Lührmann, A., Hellmeier, S., Grahn, S., and Lindberg, S. I. (2020). State
of the world 2019: Autocratization surges — resistance grows. Democratization,
27(6), 909–927.
Magaloni, B. (2006). Voting for Autocracy: Hegemonic Party Survival and Its Demise
in Mexico. New York, NY: Cambridge University Press.
Magaloni, B., Diaz-Cayeros, A., and Estévez, F. (2007). Clientelism and portfolio
diversification: A model of electoral investment with applications to Mexico.
In: H. Kitschelt and S. Wilkinson (Eds.), Patrons or Policies? Patterns of
Democratic Accountability and Political Competition, pp. 182–205. Cambridge,
Magaloni, B. and Kricheli, R. (2010). Political order and one-party rule. Annual
Maghraoui, D. (2011). Constitutional reforms in Morocco: Between consensus and
subaltern politics. Journal of North African Studies, 16(4), 679–699.
Mainwaring, S. (1986). The transition to democracy in Brazil. Journal of
Interamerican Studies and World Affairs, 28(1), 149–179.
Mainwaring, S. (1991). Politicians, parties, and electoral systems: Brazil in compar-
ative perspective. Comparative Politics, 24(1), 21–43.
Mainwaring, S. (1998). Electoral volatility in Brazil. Party Politics, 4(4), 523–545.
Mainwaring, S. (1999). Rethinking Party Systems in the Third Wave of Democratization:
The case of Brazil. Stanford, CA: Stanford University Press.
Mainwaring, S., Brinks, D., and Pérez-Liñán, A. (2001). Classifying political
regimes in Latin America, 1945–1999. Studies in Comparative International
Development, 36(1), 37–65.
Mainwaring, S., Gervasoni, C., and España-Nájera, A. (2010). The vote share of new
and young parties. Kellog Institute Working Paper WPS#368. Notre Dame, IN:
University of Notre Dame.
Mainwaring, S. and Scully, T. R. (Eds.). (1995). Building Democratic Institutions:
Party Systems in Latin America. Stanford, CA: Stanford University Press.
Mainwaring, S. and Zoco, E. (2007). Political sequences and the stabilization of
interparty competition: Electoral volatility in old and new democracies. Party
Politics, 13(2), 155–178.
Mair, P. (1989). The problem of party system change. Journal of Theoretical Politics,
1(3), 251–276.
Mair, P. (1994). Party organizations: From civil society to the state. In: R. S. Katz
and P. Mair (Eds.), How Parties Organize: Change and Adaptation in Party
Organizations in Western Democracies, pp. 1–22. London, UK: Sage.

References 253
Mair, P. (2002). Comparing party systems. In: L. LeDuc, R. G. Niemi, and P. Norris
(Eds.), Comparing Democracies 2: New Challenges in the Study of Elections and
Voting, pp. 88–107. London, UK: Sage.
Makara, S., Rakner, L., and Svåsand, L. (2009). Turnaround: The National
Resistance Movement and the reintroduction of a multiparty system in Uganda.
International Political Science Review, 30(2), 185–204.
Malesky, E. and Schuler, P. (2009). Paint-by-numbers democracy: The stakes, struc-
ture, and results of the 2007 Vietnamese National Assembly election. Journal
of Vietnamese Studies, 4(1), 1–48.
Malloy, J. and Thorn, R. (1971). Beyond the Revolution: Bolivia since 1952. Pittsburgh,
PA: University of Pittsburgh Press.
Manning, C. (2005). Assessing African party systems after the third wave. Party
Politics, 11(6), 707–727.
March, A. F. (2003). From Leninism to Karimovism: Hegemony, ideology, and
authoritarian legitimation. Post-Soviet Affairs, 19(4), 307–336.
Marcus, R. R. and Ratsimbaharison, A. M. (2005). Political parties in Madagascar:
Neopatrimonial tools or democratic instruments? Party Politics 11(4), 495–512.
Mark, S. and Trapido, S. (Eds.). (2014). The Politics of Race, Class and Nationalism
in Twentieth Century South Africa. Abingdon, UK: Routledge.
Marshall, M. G., Gurr, T. R., and Jaggers, K. (2016). Polity IV project: Political
regime characteristics and transitions, 1800–2015, dataset users’ manual.
Vienna, VA: Center for Systemic Peace. Retrieved from http://systemicpeace.
org/inscrdata.html.
Martins, H. (1969). Opposition in Portugal. Government and Opposition, 4(2),
250–263.
Martz, J. D. (1964). Dilemmas in the study of Latin American parties. Journal of
Politics, 26(8), 509–531.
Maryanov, G. S. (1967). Political parties in Mainland Malaya. Journal of Southeast
Asian History, 8(1), 99–110.
Massicotte, L. and Blais, A. (1999). Mixed electoral systems: a conceptual and
empirical survey. Electoral Studies, 18(3), 341–366.
McCoy, A. W. (1989). Quezon’s commonwealth: The emergence of Philippine
authoritarianism. In: R. R. Paredes (Ed.), Philippine Colonial Democracy,
pp. 114–160. Quezon City, Philippines: Ateneo de Manila University Press.
McKinlay, R. D. and Cohan, A. S. (1975). A comparative analysis of the political
and economic performance of military and civilian regimes: A cross-national
aggregate study. Comparative Politics, 8(1), 1–30.

6"×9"
Meguid, B. (2008). Party Competition between Unequals: Strategies and Electoral

Fortunes in Western Europe. Cambridge, UK: Cambridge University Press.
Merkel, W. (2004). Embedded and defective democracies. Democratization, 11(5),
33–58.
Miller, M. K. (2015a). Democratic pieces: Autocratic elections and democratic
development since 1815. British Journal of Political Science, 45(3), 501–530.
Miller, M. K. (2015b). Elections, information, and policy responsiveness in autocratic
regimes. Comparative Political Studies, 48(6), 691–727.
Mitchell, C. (1977). The Legacy of Populism in Bolivia: From the MNR to Military
Rule. New York, NY: Praeger.
Mobasher, M. B. (2019). Constitutional features of presidential elections and the
failure of cross-ethnic coalitions to institutionalize. In: S. Ratuva (Ed.), The
Palgrave Handbook of Ethnicity, pp. 1–20. London, UK: Palgrave Macmillan.
Moenius, J. and Kasuya, Y. (2004). Measuring party linkage across districts. Party
Politics, 10(5), 543–564.
Molinar, J. (1991). Counting the number of parties: An alternative index. American
Møller, J. and Skaaning, S.-E. (2013). The third wave: Inside the numbers. Journal
of Democracy, 24(4), 97–109.
Morgenbesser, L. (2019). Cambodia’s transition to hegemonic authoritarianism.
Morgenbesser, L. and Pepinsky, T. B. (2019). Elections as causes of democratiza-
tion: Southeast Asia in comparative perspective. Comparative Political Studies,
52(1), 3–35.
Morgenstern, S. and Swindle, S. (2005). Are politics local? An analysis of voting
patterns in 23 democracies. Comparative Political Studies, 38(2), 143–170.
Morgenstern, S., Swindle, S. M., and Castagnola, A. (2009). Party nationalization
and institutions. Journal of Politics, 71(4), 1322–1341.
Morlino, L. (2009). Are there hybrid regimes? Or are they just an optical illusion?
European Political Science Review, 1(2), 273–296.
Morrison, D. G., Cameron, R., and Paden, J. N. (1989). Congo. In: D. G. Morrison,
R. Cameron, and J. N. Paden (Eds.), Black Africa: A Comparative Handbook,
pp. 419–427. London, UK: Palgrave Macmillan.
Morse, Y. L. (2015). From single-party to electoral authoritarian regimes: The
institutional origins of competitiveness in post-Cold War Africa. Comparative
Politics, 48(1), 126–151.

References 255
Morse, Y. L. (2020). Rebuilding, rebranding, and competitive landscapes: A set-

theoretic analysis of authoritarian successor parties. Democratization,
27(8), 1477–1495.
Mortimer, R. (2006). State and army in Algeria: The ‘Bouteflika effect’. Journal of
North African Studies, 11(2), 155–171.
Moser, R. G. and Scheiner, E. (2012). Electoral Systems and Political Context: How
the Effects of Rules Vary Across New and Established Democracies. New York,
NY: Cambridge University Press.
Mozaffar, S. and Scarritt, J. (2005). The puzzle of African party systems. Party
Politics, 11(4), 399–421.
Mozaffar, S. and Schedler, A. (2002). The comparative study of electoral governance:
Introduction. International Political Science Review, 23(1), 5–27.
Mueller, D. C. and Stratmann, T. (2003). The economic effects of democratic
participation. Journal of Public Economics, 87(9–10), 2129–2155.
Mufti, M. (2018). What do we know about hybrid regimes after two decades of
scholarship? Politics and Governance, 6(2), 112–119.
Mufti, M., Shafqat, S., and Siddiqui, N. (Eds.). (2020). Pakistan’s Political Parties:
Surviving between Dictatorship and Democracy. Washington, DC: Georgetown
University Press.
Munck, G. and Verkuilen, J. (2002). Conceptualizing and measuring democracy:
Evaluating alternative indices. Comparative Political Studies, 35(1), 5–34.
Nadzri, M. M. N. (2018). The 14th general election, the fall of Barisan Nasional,
and political development in Malaysia, 1957–2018. Journal of Current Southeast
Asian Affairs, 37(3), 139–171.
Neher, C. D. (1970). Constitutionalism and elections in Thailand. Pacific Affairs,
43(2), 240–257.
Neumann, S. (1956). Towards a comparative study of political parties. In:
S. Neumann (Ed.), Modern Political Parties, pp. 395–421. Chicago, IL: Chicago
University Press.
Nickson, A. (1989). The overthrow of the Stroessner regime: Re-establishing the
status quo. Bulletin of Latin American Research, 8(2), 185–209.
Niedermayer, O. (1995). Party system change in East Germany. German Politics,
4(3), 75–91.
Niemi, R. G. and Hsieh, J. F. (2002). Counting candidates: An alternative to the
effective N (with an application to the M + 1 rule in Japan). Party Politics,
8(1), 75–99.

6"×9"
Nohlen D. (2005a). Elections in the Americas: A Data Handbook. Vol. I: North America,
Central America, and the Caribbean. Oxford, UK: Oxford University Press.
Nohlen, D. (2005b). Elections in the Americas: A Data Handbook. Vol. II: South
America. Oxford, UK: Oxford University Press.
Nohlen, D., Grotz, F., and Hartmann, C. (2001a). Elections in Asia and the Pacific:
A Data Handbook. Vol. I: Middle East, Central Asia, and South Asia. Oxford,
UK: Oxford University Press.
Nohlen, D., Grotz, F., and Hartmann, C. (2001b). Elections in Asia and the Pacific:
A Data Handbook. Vol. II: South East Asia, East Asia, and the South Pacific.
Oxford, UK: Oxford University Press.
Nohlen, D., Krennerich, M., and Thibaut, B. (1999). Elections in Africa: A Data
Handbook. Oxford, UK: Oxford University Press.
Nohlen, D. and Stöver, P. (2010). Elections in Europe: A Data Handbook. Baden-
Baden, Germany: Nomos.
Nordlinger, E. (1977). Soldiers in Politics: Military Coups and Governments.
Englewood Cliffs, NJ: Prentice-Hall.
Norris, P. (2004). Electoral Engineering: Voting Rules and Political Behaviour.
Nugent, E. R. (2020). The psychology of repression and polarization. World Politics,
72(2), 291–334.
Nwosu, B. U. (2012). Tracks of the third wave: Democracy theory, democratisation
and the dilemma of political succession in Africa. Review of African Political
Economy, 39(131), 11–25.
O’Donnell, G. (2004). Human development, human rights and democracy. In:
G. O’Donnell, O. Iazzetta, and J. Vargas Cullell (Eds.), The Quality of Democracy:
Theory and Applications, pp. 9–92. Notre Dame, IN: University of Notre
Dame Press.
O’Donnell, G. and Schmitter, P. (1986). Transitions from Authoritarian Rule: Tentative
Conclusions about Uncertain Democracies. Baltimore, MD: John Hopkins
University Press.
Olson, D. M. (1998). Party formation and party system consolidation in the new
democracies of Central Europe. Political Studies, 46(3), 432–464.
Omelicheva, M. Y. (2016). Authoritarian legitimation: Assessing discourses of
legitimacy in Kazakhstan and Uzbekistan. Central Asian Survey, 35(4), 481–450.
Ordeshook, P. C. and Shvetsova, O. V. (1994). Ethnic heterogeneity, district mag-
nitude and the number of parties. American Journal of Political Science, 38(1),
100–123.

References 257
Orlow, D. (2018). A History of Modern Germany: 1871 to Present. 7th Edition.

Ortmann, S. (2011). Singapore: Authoritarian but newly competitive. Journal of
Democracy, 22(4), 153–164.
Panebianco, A. (1988). Political Parties: Organization and Power. New York, NY:
Parajulee, R. (2010). An experiment with a hybrid regime in Nepal (1990–2006).
Journal of Asian and African Studies, 45(1), 87–112.
Patel, H. H. (1979). White power in Rhodesia: The rise and fall (?) of the Rhodesian
Front. Current Bibliography on African Affairs, 12(1), 3–18.
Pavlović, D. (2020). How political parties in hybrid regimes help the incumbent
maintain a democratic façade. East European Politics and Societies: And Cultures.
Available from http://doi.org/10.1177/0888325420941101.
Payne, S. G. (1987). The Franco Regime, 1936–1975. Madison, WI: University of
Wisconsin Press.
Pedersen, M. (1979). The dynamics of European party systems: Changing patterns
of electoral volatility. European Journal of Political Research, 7(1), 1–26.
Pepinsky, T. (2014). The institutional turn in comparative authoritarianism. British
Pérez-Liñán, A. and Mainwaring, S. (2013). Regime legacies and levels of democracy:
Evidence from Latin America. Comparative Politics, 45(4), 379–397.
Pérez-Liñán, A., Schmidt, N., and Vairo, D. (2019). Presidential hegemony and
democratic backsliding in Latin America, 1925–2016. Democratization, 26(4),
606–625.
Perthes, V. (1993). Incremental change in Syria. Current History, 92(570), 23–26.
Peters, G. (2019). Institutional Theory in Political Science: The New Institutionalism.
4th Edition. Cheltenham, UK: Edward Elgar Publishing.
Peters, G. (2020). Can we be casual about being causal? Journal of Comparative
Policy Analysis: Research and Practice. Available from: http://doi.org/10.1080
/13876988.2020.1793327.
Pop-Eleches, G. and Tucker, J. A. (2014). Communist socialization and post-
communist economic and political attitudes. Electoral Studies, 33, 77–89.
Posada-Carbó, E. (1997). Limits of power: Elections under the Conservative
hegemony in Colombia, 1886–1930. Hispanic American Historical Review,
77(2), 245–279.
Powell, G. (1982). Contemporary Democracies: Participation, Stability, and Violence.
Cambridge, MA: Harvard University Press.

6"×9"
Powell, E. N. and Tucker, J. A. (2014). Revisiting electoral volatility in post-

communist countries: New data, new results and new approaches. British
Power, T. J. and Gasiorowski, M. J. (1997). Institutional design and democratic
consolidation in the Third World. Comparative Political Studies, 30(2), 123–155.
Premdas, R. R. (1978). Guyana: Socialist reconstruction or political opportunism?
Journal of Interamerican Studies and World Affairs, 20(2), 133–164.
Pripstein Posusney, M. (2002). Multi-party elections in the Arab world: Institutional
engineering and oppositional strategies. Studies in Comparative International
Development, 36(4), 34–62.
Przeworski, A. (1991). Democracy and the Market: Political and Economic Reforms in
Eastern Europe and Latin America. New York, NY: Cambridge University Press.
Przeworski, A. (2004). Institutions matter? Government and Opposition, 39(4),
527–540.
Przeworski, A., Asadurian, T., and Bohlken, A. T. (2012). The origins of parlia
mentary responsibility. In: T. Ginsburg (Ed.), Comparative Constitutional
Design, pp. 101–137. New York, NY: Cambridge University Press.
Raby, D. L. (1989). Controlled, limited and manipulated opposition under a dic-
tatorial regime: Portugal, 1945–9. European History Quarterly, 19(1), 63–84.
Rae, D. W. (1967). The Political Consequences of Electoral Laws. New Haven, CT:
Yale University Press.
Ragauskas, R. and Thames, F. (2020). Cross-tier personal gains in mixed electoral
systems. Journal of Elections, Public Opinion and Parties. Available from: http://
doi.org/10.1080/17457289.2020.1778008.
Raymond, C. D. (2015). The organizational ecology of ethnic cleavages: The nonlin-
ear effects of ethnic diversity on party system fragmentation. Electoral Studies,
37, 109–119.
Raymond, C. D. (2018). Socioeconomic development and party system fragmen-
tation cross-nationally. Politics & Policy, 46(5), 731–758.
Rebstock, U. (2005). Democracy, Islamicity and tribalism in Mauritania. The Mario
Einaudi Center for International Studies Working Paper No. 10-05. Ithaca, NY:
Cornell University.
Reed, S. (2001). Duverger’s law is working in Italy. Comparative Political Studies,
34(3), 312–327.
Reeve, D. (1987). Gokar of Indonesia: An Alternative to the Party System. Singapore:
Oxford University Press.

References 259
Reilly, B. (2007). Electoral systems and party systems in East Asia. Journal of East
Asian Studies, 7(2), 185–202.
Reisinger, W. M. and Moraski, B. J. (2017). The Regional Roots of Russia’s Political
Regime. Ann Arbor, MI: University of Michigan Press.
Remmer, K. (1985). Redemocratization and the impact of authoritarian rule in Latin
America. Comparative Politics, 17(3), 253–275.
Reuter, O. J. (2010). The politics of dominant party formation: United Russia and
Russia’s governors. Europe-Asia Studies, 62(2), 293–327.
Reuter, O. J. (2017). The Origins of Dominant Parties: Building Authoritarian
Institutions in Post-Soviet Russia. New York, NY: Cambridge University Press.
Reynolds, A., Reilly, B., and Ellis, A. (2005). Electoral System Design: The New
International IDEA Handbook. Stockholm, Sweden: International IDEA.
Reyntjens, F. (2016). The struggle over term limits in Africa: A new look at the
evidence. Journal of Democracy, 27(3), 61–68.
Riedl, R. B. (2014). Authoritarian Origins of Democratic Party Systems in Africa.
New York, NY: Cambridge University Press.
Riedl, R. B., Slater, D., Wong, J., and Ziblatt, D. (2020). Authoritarian-led democ-
ratization. Annual Review of Political Science, 23, 315–332.
Riera, P. (2013). Electoral coordination in mixed-member systems: Does the level
of democratic experience matter? European Journal of Political Research, 52(1),
119–141.
Riera, P. (2018). The impact of intraparty electoral reforms on party system nation-
alization. Publius: The Journal of Federalism, 48(2), 191–216.
Riera, P. (2020). Socioeconomic heterogeneity and party system fragmentation.
Journal of Elections, Public Opinion and Parties. Available from: http://doi.org
/10.1080/17457289.2020.1784181.
Riker, W. H. (1975). Federalism. In: F. Greenstein and N. Polsby (Eds.), Handbook
of Political Science: Governmental Institutions and Processes, Vol. 5, pp. 93–172.
Reading, UK: Addison-Wesley.
Riker, W. H. (1982). The two-party system and Duverger’s law: An essay on the
history of political science. American Political Science Review, 76(4), 753–766.
Riley, S. and Parfitt, T. W. (1987). Party or masquerade? The All People’s Congress of
Sierra Leone. Journal of Commonwealth & Comparative Politics, 25(2), 161–179.
Rivera, S. W. (1996). Historical cleavages or transition mode? Influences on the
emerging party systems in Poland, Hungary and Czechoslovakia. Party Politics,
2(2), 177–208.

6"×9"
Roberts, D. (1994). The Cambodian elections of 1993. Electoral Studies, 13(2), 157–162.
Roberts, S. P. (2015). Converging party systems in Russia and Central Asia: A case
of authoritarian norm diffusion? Communist and Post-Communist Studies,
48(2–3), 147–157.
Roberts, T. L. (2015). The durability of presidential and parliament-based dictator-
ships. Comparative Political Studies, 48(7), 915–948.
Robinson, N. and Milne, S. (2017). Populism and political development in hybrid
regimes: Russia and the development of official populism. International Political
Science Review, 38(4), 412–425.
Rodden, J. (2009). Back to the future: Endogenous institutions and comparative
politics. In: M. I. Lichbach and A. S. Zuckerman (Eds.), Comparative Politics:
Rationality, Culture, and Structure, 2nd Edition, pp. 333–357. Cambridge, UK:
Rokkan, S. (1970). Citizens, Elections, Parties: Approaches to the Comparative Study
of Political Development. Oslo, Norway: Universitetsforlaget.
Roman, P. (1999). People’s Power: Cuba’s Experience with Representative Government.
Boulder, CO: Westview.
Rose, R. and Mackie, T. T. (1988). Do parties persist or fail? The big trade-off
facing organizations. In: K. Lawson and P. H. Merkl (Eds.), When Parties Fail:
Emerging Alternative Organizations, pp. 533–558. Princeton, NJ: Princeton
University Press.
Rose, R. and Urwin, D. W. (1970). Persistence and change in Western party systems
since 1945. Political Studies, 18(3), 287–319.
Rose, R. and Urwin, D. W. (1975). Regional Differentiation and Political Unity in
Western Nations. Sage Professional Papers in Contemporary Political Sociology,
Ser. No. 06–007. Beverly Hills, CA: Sage.
Ruhl, J. M. (2004). Curbing Central America’s militaries. Journal of Democracy, 15(3),
137–151.
Sadiki, L. (2002). Political liberalization in Bin Ali’s Tunisia: Façade democracy.
Saikkonen, I. A.-L. and White, A. C. (2020). Strategic targeting: Authoritarian
capacity, state dependent populations, and electoral manipulation. Journal of
Elections, Public Opinion and Parties. Available from: http://doi.org/10.1080/
17457289.2020.1740715.
Samuels, D. and Abrucio, F. L. (2000). Federalism and democratic transitions: The
‘new’ politics of the governors in Brazil. Publius: The Journal of Federalism,
30(2), 43–61.

References 261
Samuels, D., and Shugart, M. S. (2010). Presidents, Parties, and Prime Ministers: How
the Separation of Powers Affects Party Organization and Behavior. Cambridge,
Sanchez, O. (2009). Party non-systems: A conceptual innovation. Party Politics,
15(4), 487–520.
Sartori, G. (1970). Concept misformation in comparative politics. American Political
Science Review, 64(4), 1033–1053.
Sartori, G. (1976). Parties and Party Systems: A Framework for Analysis, Vol. 1.
Cambridge, UK: Cambridge University Press.
Sartori, G. (1986). The influence of electoral systems: Faulty laws or faulty method?
In: B. Grofman and A. Lijphart (Eds.), Electoral Laws and Their Political
Consequences, pp. 43–68. New York, NY: Agathon Press.
Sartori, G. (1987). The Theory of Democracy Revisited. Chatham, NJ: Chatham House.
Sater, J. N. (2009). Parliamentary elections and authoritarian rule in Morocco. Middle
East Journal, 63(3), 381–400.
Satpaev, D. A. (2007). An analysis of the internal structure of Kazakhstan’s political
elite and an assessment of political risk levels. In: U. Tomohiko (Ed.), Empire,
Islam, and Politics in Central Eurasia, pp. 283–300. Sapporo, Japan: Slavic
Research Center.
Scarrow, H. A. (1999). The impact of at-large elections: Vote dilution or choice
dilution? Electoral Studies, 18(4), 557–567.
Schakel, A. H. (2013). Nationalisation of multilevel party systems: A conceptual
and empirical analysis. European Journal of Political Research, 52(2), 212–236.
Schakel, A. H. (2018). Denationalization of elections? Tracing the developments
in the conceptualization and measurement of the nationalization of the vote.
In: K. Detterbeck and E. Hepburn (Eds.), Handbook of Territorial Politics,
pp. 107–119. Cheltenham, UK: Edward Elgar.
Schatz, E. (2009). The soft authoritarian tool kit: Agenda-setting power in
Kazakhstan and Kyrgyzstan. Comparative Politics, 41(2), 203–222.
Schedler, A. (2002a). The nested game of democratization by elections. International
Schedler, A. (2002b). Elections without democracy: The menu of manipulation.
Schedler, A. (2013). The Politics of Uncertainty: Sustaining and Subverting Electoral
Authoritarianism. New York, NY: Cambridge University Press.
Schirmer, J. (1998). The Guatemalan Military Project: A Violence Called Democracy.
Philadelphia, PA: University of Pennsylvania Press.

6"×9"
Schlesinger, J. A. (1994). Political Parties and the Winning of Office. Ann Arbor, MI:
University of Michigan Press.
Schofield, N. and Zakharov, A. (2010). A stochastic model of the 2007 Russian
Duma election. Public Choice, 142(1–2), 177–194.
Scigliano, R. G. (1960). The electoral process in South Vietnam: Politics in an under-
developed state. Midwest Journal of Political Science, 4(2), 138–161.
Scott, J. C. (1972). Patron-client politics and political change in Southeast Asia.
American Political Science Review, 66(1), 91–113.
Seeberg, M. B. (2014). State capacity and the paradox of authoritarian elections.
Sejjaaka, S. (2004). A political and economic history of Uganda, 1962–2002. In:
F. Bird and S. W. Herman (Eds.), International Businesses and the Challenges
of Poverty in the Developing World: Case Studies on Global Responsibilities and
Practices, pp. 98–110. London, UK: Palgrave Macmillan.
Senigaglia, C. (2020). The debate on democratization and parliament in
Germany from 1871 to 1918. Parliaments, Estates and Representation, 40(3),
290–307.
Shefter, M. (1994). Political Parties and the State: The American Historical Experience.
Princeton, NJ: Princeton University Press.
Shevchenko, I. and Golosov, G. V. (2001). Legislative activism of Russian Duma
deputies, 1996–1999. Europe-Asia Studies, 53(2), 239–261.
Shugart, M. S. (1995). The electoral cycle and institutional sources of divided
presidential government. American Political Science Review, 89(2), 327–343.
Shugart, M. S. (2005). Semi-presidential systems: Dual executive and mixed authority
patterns. French Politics, 3(3), 323–351.
Shugart, M. S. and Carey, J. M. (1992). Presidents and Assemblies: Constitutional
Design and Electoral Dynamics. New York, NY: Cambridge University Press.
Shugart, M. S. and Wattenberg, M. P. (Eds.). (2001). Mixed-Member Electoral Systems:
The Best of Both Worlds? Oxford, UK: Oxford University Press.
Siaroff, A. (2003). Comparative presidencies: The inadequacy of the presidential,
semi-presidential and parliamentary distinction. European Journal of Political
Research, 42(3), 287–312.
Siaroff, A. (2019). Comparative European Party Systems: An Analysis of Parliamentary
Elections since 1945. 2nd Edition. New York, NY: Taylor and Francis.
Sikk, A. (2005). How unstable? Volatility and the genuinely new parties in Eastern
Europe. European Journal of Political Research, 44(3), 391–412.

References 263
Sikk, A. and Taagepera, R. (2014). How population size affects party systems and
cabinet duration. Party Politics, 20(4), 591–603.
Simón, P. and Guinjoan, M. (2018). The short-term and long-term effects of institu-
tional reforms on party system nationalization. Comparative European Politics,
16(5), 762–782.
Simpser, A. (2013). Why Governments and Parties Manipulate Elections: Theory,
Practice, and Implications. New York, NY: Cambridge University Press.
Singer, M. M. and Stephenson, L. B. (2009). The political context and Duverger’s
theory: Evidence at the district level. Electoral Studies, 28(3), 480–491.
Singh, C. (2008). Re-democratization in Guyana and Suriname: Critical compari-
sons. European Review of Latin American and Caribbean Studies/Revista Europea
de Estudios Latinoamericanos y del Caribe, 84, 71–85.
Singh, M. (2017). Parliamentary election in Jordan, 2016. Contemporary Review of
the Middle East, 4(3), 297–318.
Sinpeng, A. (2014). Party banning and the impact on party system institutionaliza-
tion in Thailand. Contemporary Southeast Asia: A Journal of International and
Strategic Affairs, 36(3), 442–466.
Sirojjon, B. (2015). Multiparty system in Uzbekistan, 1985–2014. Himalayan and
Central Asian Studies, 9(3–4), 150–157.
Skaaning, S.-E., Gerring, J., and Bartusevičius, H. (2015). A Lexical index of electoral
democracy. Comparative Political Studies, 48(12), 1491–1525.
Skidmore, T. E. (1988). The Politics of Military Rule in Brazil, 1964–1985. New York,
NY: Oxford University Press.
Skovoroda, R. and Lankina, T. (2017). Fabricating votes for Putin: New tests of
fraud and electoral manipulation from Russia. Post-Soviet Affairs, 33(2),
100–123.
Slater, D. (2003). Iron cage in an iron fist: Authoritarian institutions and the per-
sonalization of power in Malaysia. Comparative Politics, 36(1), 81–101.
Smith, G. (1989). A system perspective on party system change. Journal of Theoretical
Politics, 1(3), 349–364.
Smith, P. H. (2005). Democracy in Latin America: Political Change in Comparative
Perspective. New York, NY: Oxford University Press.
Spence, J. E. (1991). The South African political process. Government and Opposition,
26(3), 316–328.
Sperber, J. (2003). Comments on Marcus Kreuzer’s article. Central European History,
36(3), 359–366.

6"×9"
Stanley, W. D. (2006). El Salvador: State-building before and after democratisation,

1980–95. Third World Quarterly, 27(1), 101–114.
Starodubtsev, A. (2018). Federalism and Regional Policy in Contemporary Russia.
Stepan, A. (1988). Rethinking Military Politics: Brazil and the Southern Cone.
Princeton, NJ: Princeton University Press.
Stepan, A. (2001). Toward a new comparative politics of federalism, (multi)
nationalism, and democracy. In: A. Stepan (Ed.), Arguing Comparative Politics,
pp. 315–361. Oxford, UK: Oxford University Press.
Stephens, E. H. (1983). The Peruvian military government, labor mobilization,
and the political strength of the left. Latin American Research Review, 18(2),
57–93.
Stockemer, D. (Ed.). (2019). Populism around the World: A Comparative Perspective.
Cham, Switzerland: Springer.
Stokes, D. E. (1967). Parties and the nationalization of electoral forces. In: W. N.
Chambers and W. D. Burnham (Eds.), The American Party Systems: States
of Political Development, pp. 182–202. New York, NY: Oxford University
Press.
Stokes, S. C. (2000). Perverse accountability: A formal model of machine politics
with evidence from Argentina. American Political Science Review, 99(3),
315–325.
Stokes, S. C. (2007). Political clientelism. In: C. Boix and S. C. Stokes (Eds.), Oxford
Handbook of Comparative Politics, pp. 604–627. Oxford, UK: Oxford University
Press.
Stoll, H. (2013). Changing Societies, Changing Party Systems. New York, NY:
Stoll, H. (2015). Presidential coattails: A closer look. Party Politics. 21(3),
417–427.
Stonecash, J. M. and Brewer, M. (2009). The Dynamics of American Political Parties.
New York, NY: Cambridge University Press.
Stroh, A. (2010). Electoral rules of the authoritarian game: Undemocratic effects
of proportional representation in Rwanda. Journal of Eastern African Studies,
4(1), 1–19.
Su, Y.-P. (2018). Personal vote, spatial registration rules, and party system
nationalization in Latin America. International Political Science Review, 39(2),
192–208.

References 265
Sultanova, S. (2014). Challenging the Aliyev regime: Political opposition in

Azerbaijan. Demokratizatsiya: The Journal of Post-Soviet Democratization,
22(1), 15–37.
Suryadinata, L. (1997). Golkar of Indonesia: Recent developments. Contemporary
Southeast Asia, 19(2), 190–204.
Svolik, M. W. (2012). The Politics of Authoritarian Rule. Cambridge, UK: Cambridge
University Press.
Svolik, M. W. (2015). Which democracies will last? Coups, incumbent takeovers,
and the dynamic of democratic consolidation. British Journal of Political Science,
45(4), 715–738.
Taagepera, R. (1999). Supplementing the effective number of parties. Electoral
Studies, 18(4), 497–504.
Taagepera, R. (2002). Limiting Frames of Political Games: Logical Quantitative
Models of Size, Growth and Distribution. Irvine, CA: Center for the Study of
Democracy (University of California, Irvine), Paper 02’04.
Taagepera, R. and Allik, M. (2006). Seat share distribution of parties: Models and
empirical patterns. Electoral Studies, 25(4), 696–713.
Taagepera, R. and Grofman, B. (1981). Effective size and number of components.
Sociological Methods and Research, 10(1), 63–81.
Taagepera, R. and Grofman, B. (1985). Rethinking Duverger’s law: Predicting the
effective number of parties in plurality and PR systems — Parties minus issues
equals one. European Journal of Political Research, 13(4), 341–352.
Taagepera, R. and Grofman, B. (2003). Mapping the indices of seats-votes dispro-
portionality and inter-election volatility. Party Politics, 9(6), 659–677.
Taagepera, R. and Shugart, M. S. (1989). Seats and Votes: The Effects and Determinants
of Electoral Systems. New Haven, CT: Yale University Press.
Tan, N. (2014). Institutionalized succession and hegemonic party cohesion
in Singapore. In: A. Hicken and E. M. Kuhonta (Eds.), Party System
Institutionalization in Asia: Democracies, Autocracies, and the Shadows of the
Past, pp. 49–73. New York, NY: Cambridge University Press.
Tavits, M. (2006). Party system change: Testing a model of new party entry. Party
Politics, 12(1), 99–119.
Taylor, M. (2019). Every citizen a soldier: The Guyana people’s militia, 1976–1985.
Journal of Global South Studies, 36(2), 279–311.
Tejapira, K. (2016). The irony of democratization and the decline of royal hegemony
in Thailand. Southeast Asian Studies, 5(2), 219–237.

6"×9"
Teorell, J. and Lindstedt, C. (2010). Measuring electoral systems. Political Research

Quarterly, 63(2), 434–448.
Tertytchnaya, K. T. (2020). Protests and voter defections in electoral autocracies:
Evidence from Russia. Comparative Political Studies, 53(12), 1926–1956.
Thames, F. C. and Edwards, M. S. (2006). Differentiating mixed-member electoral
systems: Mixed-member majoritarian and mixed-member proportional systems
and government expenditures. Comparative Political Studies, 39(7), 905–927.
Thorlakson, L. (2007). An institutional explanation of party system congruence:
Evidence from six federations. European Journal of Political Research, 46(1),
69–95.
Tien, H. (Ed.). (1996). Taiwan’s Electoral Politics and Democratic Transition: Riding
the Third Wave. Armonk, NY: M. E. Sharpe.
Tkacheva, T. and Golosov, G. V. (2019). United Russia’s primaries and the strength
of political machines in the regions of Russia: Evidence from the 2016 Duma
elections. Europe-Asia Studies, 71(5), 824–839.
Tóka, G. (1998). Party appeals and voter loyalty in new democracies. Political
Studies, 46(3), 589–610.
Tóka, G. (2009). Expressive vs. instrumental motivation of turnout, partisanship
and political learning. In: H.-D. Klingemann (Ed.), The Comparative Study of
Electoral Systems, pp. 269–288. Oxford: Oxford University Press.
Totman, S. and Hardy, M. (2015). When good dictators go bad: Examining the
‘transformation’ of Colonel Gaddafi. International Journal of Interdisciplinary
Global Studies, 10(2), 1–6.
Treier, S. and Jackman, S. (2008). Democracy as a latent variable. American Journal
of Political Science, 52(1), 201–217.
Tucker, J. A. (2007). Enough! Electoral fraud, collective action problems, and
post-communist colored revolutions. Perspectives on Politics, 5(3), 535–551.
Tullock, G. (1971). The charity of the uncharitable. Western Economic Journal, 9(4),
379–392.
Turchenko, M. (2020). Electoral engineering in the Russian regions (2003–2017).
Europe-Asia Studies, 72(1), 80–98.
Turchenko, M. and Golosov, G. V. (2021). Smart enough to make a difference? An
empirical test of the efficacy of strategic voting in Russia’s authoritarian elections.
Post-Soviet Affairs, 37(1), 65–79.
Ufen, A. (2012). Party systems, critical junctures, and cleavages in Southeast Asia.
Asian Survey, 52(3), 441–464.

References 267
Ufen, A. (2020). Opposition in transition: Pre-electoral coalitions and the 2018

electoral breakthrough in Malaysia. Democratization, 27(2), 167–184.
Ulfelder, J. and Lustik, M. (2007). Modelling transitions to and from democracy.
Valenzuela, J., Somma, N., and Scully, T. (2018). Resilience and change: The party
system in re-democratized Chile. In: S. Mainwaring (Ed.), Party Systems in Latin
America: Institutionalization, Decay, and Collapse, pp. 135–163. Cambridge,
Van der Kolk, H. (2007). Local electoral systems in Western Europe. Local
Government Studies, 33(2), 159–180.
Van Dijk, J. W. M. (2007). Political deadlock in Chad. African Affairs, 106(425),
697–703.
Van Ham, C. and Lindberg, S. I. (2016). Choosing from the menu of manipulation:
Explaining incumbents’ choices of electoral manipulation tactics. V-Dem
Institute Working Paper 30, University of Gothenburg, Gothenburg.
Vandeginste, S. (2014). Governing ethnicity after genocide: Ethnic amnesia in
Rwanda versus ethnic power-sharing in Burund. Journal of Eastern African
Studies, 8(2), 263–277.
Vanhanen, T. (2000). A new dataset for measuring democracy, 1810–1998. Journal
of Peace Research, 37(2), 251–265.
Vertz, L. L., Frendreis, J. P., and Gibson J. L. (1987). Nationalization of the electorate
in the United States. American Political Science Review, 81(3), 961–966.
Vladisavljević, N. (2016). Competitive authoritarianism and popular protest:
Evidence from Serbia under Milošević. International Political Science Review,
37(1), 36–50.
Völkel, J. C. (2017). Sidelined by design: Egypt’s parliament in transition. Journal
of North African Studies, 22(4), 595–619.
Volpi, F. (2020). Algeria: When elections hurt democracy. Journal of Democracy,
31(2), 152–165.
Von Beyme, K. (2000). Parliamentary Democracy: Democratization, Destabilization,
Reconsolidation, 1789–1999. New York, NY: St. Martin’s Press.
Wahman, M. (2011). Offices and policies — Why do oppositional parties form
pre-electoral coalitions in competitive authoritarian regimes? Electoral Studies,
30(4), 642–657.
Wahman, M. (2013). Opposition coalitions and democratization by election.
Government and Opposition, 48(1), 3–32.

6"×9"
Wang, Y., Mechkova, V., and Andersson, F. (2019). Does democracy enhance
health? New empirical evidence 1900–2012. Political Research Quarterly, 72(3),
554–569.
Ward, P. M. and Rodriguez, V. E. (1999). New federalism, intra-governmental
relations and co-governance in Mexico. Journal of Latin American Studies,
31(3), 673–710.
Ware, A. (1995). Political Parties and Party Systems. Oxford, UK: Oxford University
Press.
Warren-Boulton, F. R. (1990). Implications of U. S. experience with horizontal
mergers and takeovers for Canadian competition policy. In: F. Mathewson,
M. Trebilcock and M. Walker (Eds.), The Law and Economics of Competition
Policy. Vancouver, Canada: The Fraser Institute.
Weber, T. (2011). Exit, voice, and cyclicality: A micrologic of midterm effects in
European Parliament elections. American Journal of Political Science, 55(4),
906–921.
Weiss, V. L. (2004). The antidemocratic potential of party system institutionalization:
Malaysia as morality tale? In: A. Hicken and E. M. Kuhonta (Eds.), Party System
Institutionalization in Asia: Democracies, Autocracies, and the Shadows of the
Past, pp. 25–48. New York, NY: Cambridge University Press.
Westwood, A. F. (1961). Elections and politics in Iran. Middle East Journal, 15(2),
153–164.
White, S. and Mcallister, I. (2008). The Putin phenomenon. Journal of Communist
Studies and Transition Politics, 24(4), 604–628.
Wigell, M. (2008). Mapping ‘hybrid regimes’: Regime types and concepts in com-
parative politics. Democratization, 15(2), 230–250.
Williams-Wyche, S. N. (2014). An empirical test of presidentialism’s effect on party
competition. Electoral Studies, 33, 166–174.
Wilson, K. (2016). How increased competition can strengthen electoral authori-
tarianism: Party-system pluralization in Russia. Problems of Post-Communism,
63(4), 199–209.
Wilson, M. C. and Piazza, J. A. (2013). Autocracies and terrorism: Conditioning
effects of authoritarian regime type on terrorist attacks. American Journal of
Wintrobe, R. (1998). The Political Economy of Dictatorship. Cambridge, UK:

References 269
Wolff, H., Chong, H., and Auffhammer, M. (2011). Classification, detection and
consequences of data error: Evidence from the Human Development Index.
Economic Journal, 121(553), 843–870.
Wong, S. H.-W. and Or, N. H. K. (2020). To compete or to cooperate: Intra-elite
dynamics in an electoral autocracy. Communist and Post-Communist Studies,
53(4), 91–117.
Woodward, C. A. (1980). One party dominance in democracy: Widening the per-
spective on the regnancy of the National Party in South Africa. Politikon: South
African Journal of Political Studies, 7(2), 111–125.
Wright, J. (2008). Do authoritarian institutions constrain? How legislatures impact
economic growth and foreign aid effectiveness. American Journal of Political
Science, 52(2), 322–342.
Wu, Y.-S. (2011). Clustering of semi-presidentialism: A first cut. In: R. Elgie,
S. Moestrup, and Y.-S. Wu (Eds.), Semi-Presidentialism and Democracy,
pp. 21–41. New York, NY: Palgrave Macmillan.
Zaccara, L. (2013). Comparing elections in Gulf Cooperation Council countries
after the Arab Spring: The United Arab Emirates, Oman, and Kuwait. Journal
of Arabian Studies, 3(1), 80–101.
Zagarri, R. (1987). The Politics of Size. Ithaca, NY: Cornell University Press.
Zartman, I. W. (1967). Political pluralism in Morocco. Government and Opposition,
2(4), 568–583.
Zaslavsky, V. and Brym, R. J. (1978). The functions of elections in the USSR. Soviet
Studies, 30(3), 362–371.

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Index
A Belgium, 31–32, 63
Afghanistan, 20 Belize, 63
Algeria, 24, 46, 95, 100, 140, 176 Benin, 45
Aliyev, Heydar, 48 Birendra Bir Bikram Shah, 42
Aliyev, Ilham, 48 Biya, Paul, 109
Angola, 24, 45, 96, 100, 140, 176 Bolivia, 24, 38, 63, 96, 100, 140, 176,
Argentina, 63 199
Armenia, 24, 48, 97, 100, 114–115, Bouteflika, Abdelaziz, 46
140, 176, 199 Brazil, 24, 37, 63, 95, 100, 105, 107,
Australia, 63 110–111, 140, 145–146, 161, 173,
Austria, 32, 63, 137 176, 199
Austro-Hungary, 31 Brunei, 139
authoritarian regimes, types of, 11–12, Bulgaria, 63
94 Burnham, Forbes, 97
authoritarianism, 1–26, 90–96 Burundi, 24, 47, 96, 100, 140, 176,
defined substantively, 8, 13 184, 189
defined operationally, 16–17
autocratization, 211–212 C
Azerbaijan, 24, 48, 99–100, 140, 176 Cabo Verde, 63
Cambodia, 11, 24, 42, 94, 99–100,
B 140, 176
Bahamas, 63 Cameroon, 24, 45, 96, 100, 109, 140,
Balaguer, Joaquín, 36 176, 185
Bangladesh, 210–211 Canada, 63
Barbados, 63 Chad, 24, 47, 96, 100, 140, 176, 185
Belarus, 20 Chile, 36, 63
271
b4407_Index.indd 271 1/6/2022 9:16:59 AM

6"×9"
China, 21, 39 electoral authoritarianism, defined, 13

clientelism, 113–115 electoral democracy index, defined,
Colombia, 21, 24, 33, 99–100, 140, 16
176, 198–199 electoral malpractice, 10
Congo, Democratic Republic of the, electoral systems, 169–193
150 defined, 169
Congo, Republic of, 24, 46, 92–93, 95, Equatorial Guinea, 24, 45, 100, 140,
100, 140, 176 177
Costa Rica, 63 Estonia, 63
Cote d’Ivoire, 24, 176, 199 Ethiopia, 24, 46, 96, 100, 140, 142,
Croatia, 63 161–162, 177
Cuba, 19, 97 executive–legislative relations,
Cyprus, 63 135–160
Czech Republic, 63 types of, 135–137
D F
Déby, Idriss, 47 federalism, 160–167
democracy, defined, 8 defined, 160
Denmark, 31–32, 63 Finland, 63, 137
Djibouti, 24, 45, 96, 100, 140, 147, format (of party systems), 66–76
176, 185 defined substantively, 66
Dominica, 63 defined operationally, 70
Dominican Republic, 24, 36, 63, 95, fragmentation (of party systems),
100, 140, 177, 199 53–66
defined substantively, 53
E defined operationally, 59
effective number of parties see France, 42, 63, 137
fragmentation (of party systems),
defined operationally G
Egypt, 24, 44, 96, 100, 140, 177, 199, Gabon, 24, 45, 85, 96, 100, 140, 177,
209 199
El Salvador, 23–24, 36, 63, 95, 100, Gaddafi, Muammar, 13
106, 140, 147, 177, 199 Gambia, 24, 46, 95, 100, 140, 177, 199
b4407_Index.indd 272 1/6/2022 9:16:59 AM

Index 273
Germany, 21, 30–32, 95 K

Ghana, 45, 63 Kabila, Joseph, 150
Greece, 63 Karimov, Islam, 50, 147
Grenada, 63 Kazakhstan, 24, 49, 99–100, 140, 177,
Guatemala, 24, 37, 85, 95, 100, 182, 189
105–106, 140, 147, 177, 199 Kosovo, 48
Guinea, 211
Guyana, 24, 38, 63, 97, 100, 110, 115, L
140, 146, 177, 199 Libya, 13, 134
Gyanendra Shah, 104 Liechtenstein, 63
Luxembourg, 63
H
Hasan II, 44 M
Horthy, Miklós, 32 Madagascar, 24, 43, 98, 100, 140, 177,
Hungary, 32, 63 199
majoritarian electoral systems, types
I of, 172–173
Iceland, 63, 137 Malaysia, 24, 27, 39–40, 94, 98, 100,
India, 57, 63 140, 142, 144–145, 161, 163, 177,
Indonesia, 24, 40, 63, 90, 92, 95, 100, 199
105, 107, 110, 140, 145–146, 177, Malta, 63
199 Mauritania, 24, 46, 95, 100, 105, 111,
Iran, 24, 41–42, 90, 94, 100, 104, 111, 140, 177, 185, 199
134–135, 140, 177, 199, 201 Mauritius, 63, 180
Iraq, 21 Medvedev, Dmitry, 149–150
Ireland, 63, 137 Mexico, 24, 27, 34–35, 37, 63, 99–100,
Israel, 63 114–115, 140, 161–162, 178, 198–199
Italy, 32, 63, 96 Micronesia, Federated States of, 137
military regime, defined, 95
J Milosevic, Slobodan, 48
Jamaica, 63 Mirziyoyev, Shavkat, 147
Japan, 32, 63 mixed electoral systems, definition
Jordan, 20, 209 and types of, 174
b4407_Index.indd 273 1/6/2022 9:16:59 AM

6"×9"
Mohamad, Mahathir, 144 P

Mohammad Reza Pahlavi, 41 Pakistan, 22, 210–211
monarchy, defined, 94–95 Panama, 63
Mongolia, 63 Paraguay, 25, 36, 96, 100, 140, 178,
Morocco, 24, 44–45, 94, 100, 103–104, 182, 199
106, 111, 135, 139–140, 178 Park Chung Hee, 39
Mosaddegh, Mohammad, 111 parliamentary system, defined, 136
Mugabe, Robert, 43 party electoral authoritarian regime,
Museveni, Yoweri, 96–97 defined, 97
Myanmar, 104, 209 party system properties, 51–87
defined, 51
N party system, defined, 4, 18
Namibia, 63 Pedersen index see Volatility
nationalization (of party systems), (of party systems), defined
80–87 operationally
defined substantively, 80 Peralta Azurdia, Enrique, 37, 106
defined operationally, 83 personalistic regimes, definition and
Nazarbayev, Nursultan, 49 varieties of, 90, 94
Nepal, 20, 25, 42, 94, 100, 104, 111, Peru, 36, 64
135, 140, 178, 199 Philippines, 25, 34, 64, 98, 100, 140,
Netherlands, 63 178, 199
New Zealand, 63 Pinochet, Augusto, 36
Ngo Dinh Diem, 39 Poland, 32, 64
Ngouabi, Marien, 92–93 political parties, defined, 17–18
Nicaragua, 25, 33–34, 36, 95, 100, 140, political regime, 7–8, 10–12, 89–132,
178, 182, 198–199, 211 139–141, 197–198, 201–215
non-party electoral authoritarian defined, 8
regime, defined, 97 Portugal, 25, 28, 33, 64, 96, 98, 100,
North Korea, 21 137, 140, 178, 199
Norway, 63 post-democracy, defined, 94
presidential system, defined, 136
O proportional representation systems,
Oman, 139 types of, 173–174
Orange Free State, 19 Putin, Vladimir, 49, 97, 149–150
b4407_Index.indd 274 1/6/2022 9:16:59 AM

Index 275
R Slovenia, 64
Rahmon, Emomali, 49–50 Somoza family, 34, 36
regime genomes, 90–132 South Africa, 25, 35, 64, 98, 100, 140,
defined, 6, 93–94 146, 178, 199
relative-size triangle see format South Korea, 25, 38–39, 64, 95, 100,
(of party systems), defined 140, 178, 199
operationally South Vietnam, 25, 38–39, 99–100,
Romania, 63 140, 179, 199
Russia, 25, 31, 48–49, 97, 100, 110, Soviet Union, 1, 27, 32, 48, 99, 196
114–115, 140, 142, 149, 161–163, Spain, 33, 64, 96
178, 180, 182–184, 189, 213 St. Kitts and Nevis, 64
Rwanda, 25, 47, 96, 100, 140, 178, 184, St. Lucia, 64
189 St. Vincent and Grenadines, 64
Stevens, Siaka, 99
S Stroessner, Alfredo, 36
Salazar, António, 33 Sudan, 25, 46, 95, 100, 140, 161, 179,
Saleh, Ali Abdullah, 40 199, 201
Samoa, 64 Suharto, 40, 90–92, 145
San Marino, 64, 137 Suriname, 64, 137
Sao Tome and Principe, 64 Sweden, 64
Sarney, José, 145 Switzerland, 64, 137
Sassou Nguesso, Denis, 93 Syria, 21
semi-presidential system, defined, system with parliamentary elected
137 executive president, defined,
Senegal, 64 137
Serbia, 25, 48, 64, 96, 100, 140, 178,
199 T
Seychelles, 25, 45, 96, 100, 140, 178, Taiwan, 25, 38–39, 64, 96, 100, 140,
185, 198–199 179, 200
Sierra Leone, 25, 43, 64, 94, 98, 100, Tajikistan, 25, 49, 99–100, 140, 179
140, 178, 199 Thailand, 25, 41, 94–95, 100, 104,
Singapore, 25, 39–40, 98, 100, 140, 111, 135, 139–140, 179, 200,
142, 178 210–211
single-party regime, defined, 95–96 Togo, 25, 45, 96, 100, 140, 179, 200
b4407_Index.indd 275 1/6/2022 9:16:59 AM

6"×9"
Trinidad and Tobago, 64 V

Trujillo family, 36 Velasco Alvarado, Juan, 36
Tshisekedi, Félix, 150 Venezuela, 209
Tsiranana, Philibert, 43 volatility (of party systems), 76–80
Tunisia, 25, 44, 96, 100, 140, 179, 182, defined substantively, 76
185, 200 defined operationally, 77
Turkey, 211, 213 voting behavior, 112–113
Turkmenistan, 210–211
Y
U Yeltsin, Boris, 110
Uganda, 25, 47, 96–97, 100, 140, 179 Yemen, 25, 40, 95, 100, 140, 179,
United Kingdom, 31–32, 42, 64 200–201
United States, 30, 64, 80, 98, 114
Uruguay, 64 Z
Uzbekistan, 25, 50, 99–100, 111, 140, Zimbabwe, 25–26, 43, 98, 100, 140,
147, 179 179, 185, 200
b4407_Index.indd 276 1/6/2022 9:16:59 AM

Authoritarian Party SystemsParty Politics in Autocratic Regimes PDF

Uploaded by

Copyright:

Available Formats

Authoritarian Party SystemsParty Politics in Autocratic Regimes PDF

Uploaded by

Document Information

Original Title

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

Authoritarian Party SystemsParty Politics in Autocratic Regimes PDF

Uploaded by

Copyright:

Available Formats

What is the main topic discussed in the document?

What is the main topic discussed in the document?

What are some examples of authoritarian party systems discussed?

What are some examples of authoritarian party systems discussed?

6"×9" b4407 Authoritarian Party Systems

Authoritarian Party Systems

Grigorii V Golosov is Professor of Comparative Politics and Head of Political

b4407_FM.indd 1 1/6/2022 9:16:43 AM

This page intentionally left blank

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

This page intentionally left blank

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

Library of Congress Cataloging-in-Publication Data

British Library Cataloguing-in-Publication Data

Copyright © 2022 by World Scientific Publishing Europe Ltd.

For any available supplementary material, please visit

Desk Editors: Nimal Koliyat/Michael Beale/Shi Ying Koe

Typeset by Stallion Press

Nimal Koliyat - Q0330 - Authoritarian Party Systems.indd 1 23/12/2021 4:09:56 pm

This book is dedicated to my beloved wife, Iulia Shevchenko.

b4407_FM.indd 7 1/6/2022 9:16:43 AM

This page intentionally left blank

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

b4407_FM.indd 9 1/6/2022 9:16:43 AM

x Authoritarian Party Systems

b4407_FM.indd 10 1/6/2022 9:16:43 AM

List of Figures xiii

Chapter 2 Defining Authoritarian Party Systems 7

Chapter 3 The World of Authoritarian Party Systems 27

Chapter 4 The Properties of Authoritarian Party Systems 51

b4407_FM.indd 11 1/6/2022 9:16:43 AM

xii Authoritarian Party Systems

Chapter 5 Political Regimes and Authoritarian Party Systems 89

Chapter 6 Authoritarian Party Systems, Executive–Legislative

Chapter 7 Authoritarian Party Systems and Electoral Systems 169

Chapter 8 The Destinations of Authoritarian Party Systems 195

List of the Names of Parties and Political Organizations

b4407_FM.indd 12 1/6/2022 9:16:43 AM

Figure 3.1. The historical dynamics of authoritarian party systems:

b4407_FM.indd 13 1/6/2022 9:16:43 AM

xiv Authoritarian Party Systems

Figure 7.1. The graphic representations of party constellations in

b4407_FM.indd 14 1/6/2022 9:16:43 AM

Table 2.1. Authoritarian party systems of the world, 1945–2019 24

b4407_FM.indd 15 1/6/2022 9:16:43 AM

xvi Authoritarian Party Systems

Table 4.12. Bonferroni corrected comparisons of democratic

b4407_FM.indd 16 1/6/2022 9:16:43 AM

List of Tables xvii

Table 5.16. Multilevel mixed-effects regression results, volatility

b4407_FM.indd 17 1/6/2022 9:16:43 AM

xviii Authoritarian Party Systems

Table 6.13. Multilevel mixed-effects regression results, volatility

b4407_FM.indd 18 1/6/2022 9:16:43 AM

List of Tables xix

Table 7.7. Two-way analysis of variance: electoral systems

b4407_FM.indd 19 1/6/2022 9:16:43 AM

This page intentionally left blank

B1948_1-Aoki.indd 6 9/22/2014 4:24:57 PM

6"×9" b4407 Authoritarian Party Systems

List of Figures xiii

Chapter 2 Defining Authoritarian Party Systems 7

Chapter 3 The World of Authoritarian Party Systems 27

Chapter 4 The Properties of Authoritarian Party Systems 51

Chapter 5 Political Regimes and Authoritarian Party Systems 89

Chapter 6 Authoritarian Party Systems, Executive–Legislative

Chapter 7 Authoritarian Party Systems and Electoral Systems 169

Chapter 8 The Destinations of Authoritarian Party Systems 195

Table 2.1. Authoritarian party systems of the world, 1945–2019 24

as Zimbabwe (Rhodesia), and the post-segregation system, f urther referred

Table 3.1. The geographical and chronological distribution of authoritarian