Professional Documents
Culture Documents
Selena Grimaldi
Palgrave Studies in Presidential Politics
Series Editor
Gianluca Passarelli, Sapienza University of Rome, Rome, Italy
Palgrave Studies in Presidential Politics publishes books on all aspects
of presidential politics. We welcome proposals for monographs, edited
volumes and Pivots on topics such as:
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer
Nature Switzerland AG 2023
This work is subject to copyright. All rights are solely and exclusively licensed by the
Publisher, whether the whole or part of the material is concerned, specifically the rights
of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilms or in any other physical way, and transmission or information storage and
retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc.
in this publication does not imply, even in the absence of a specific statement, that such
names are exempt from the relevant protective laws and regulations and therefore free for
general use.
The publisher, the authors, and the editors are safe to assume that the advice and informa-
tion in this book are believed to be true and accurate at the date of publication. Neither
the publisher nor the authors or the editors give a warranty, expressed or implied, with
respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
This Palgrave Macmillan imprint is published by the registered company Springer Nature
Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To my little Elettra who might consider becoming a President
rather than a Princess
With love Mum
Acknowledgements
vii
viii ACKNOWLEDGEMENTS
I’m also grateful to colleagues and friends who gave valuable remarks
and comments on certain parts of the book previously proposed as
conference papers especially in ECPR General Conferences in 2020
and 2021 and in the international workshop organized in Tampere
in 2022 and therefore to Raluca Farcas, Philip Köker, Vesa Koskimaa,
Mažvydas Jastramskis, Sophia Moestrup, Mara Morini, Octavio Amorin
Neto, Gianluca Passarelli, Lukas Pukelis and Thomas Sedelius.
I would also thank Matteo Boldrini and Aldo Paparo for the helpful
discussions on certain methodological strategies and my friends Laura and
Giulia Doriguzzi Bozzo for their helpful comments as outsider readers.
Finally, I am grateful to my parents and my partner Pierluigi Corri-
dolo for all the support and help in making worked the difficult balance
between private and working life in a period of great changes and stressful
situations.
Last but not least, thanks to Hemapriya Eswanth and Ambra Finotelli
of Palgrave McMillan for their patience and kindness and for never giving
up with the project of this book.
Contents
1 Introduction 1
1.1 What Are Weak Presidents? 5
1.2 Why Study Weak Presidents? 11
1.3 Aim, Research Questions and Theoretical Framework 16
1.4 Research Design, Methods and Data 22
References 29
2 Weak Presidents’ Characteristics and Pathways to Power 37
2.1 A Gap to Be Filled: How to Tackle with Presidents’
Background and Career 37
2.2 Presidents’ Socio-demographic Background 43
2.3 Presidents’ Political Career Paths 54
2.4 Career Duration, Diversification and Types of Presidents 69
2.5 Conclusions 79
References 81
3 Election and Term of Weak Presidents 91
3.1 Presidential Electoral Rules and Presidential Terms 91
3.2 Direct Elections and Second Order Theory 100
3.3 Indirect Elections and Party Politics 122
3.4 Duration in Office and Early Resignations 145
3.5 Conclusion 152
References 156
ix
x CONTENTS
xiii
xiv LIST OF FIGURES
xix
xx LIST OF TABLES
Table 3.15 Vote share in the final round and number of rounds
of voting per president by country (percentage
and absolute values) 142
Table 3.16 Votes of the winning candidates and the mean number
of rounds of voting per party family (mean values) 145
Table 3.17 Average duration in office during 1st and 2nd term
of directly and indirectly elected presidents and amount
of term completed 148
Table 4.1 Measures of president’s formal powers 170
Table 4.2 Presidential formal powers in government formation
and parliament dissolution in Western Europe 178
Table 4.3 Presidential formal powers in policy-making 206
Table 4.4 Referrals to the supreme court by Irish presidents
(1937–2021) 215
Table 4.5 Suspensive Vetoes of Italian presidents (1948–2021) 219
Table 4.6 Vetoes of Portuguese presidents 223
Table 5.1 Differences between formal and informal presidential
powers 241
Table 5.2 A typology of informal powers 248
Table 5.3 Mapping the proposals of governing formula
as substitutive or subverting informal powers and their
outcomes 259
Table 5.4 Mapping presidential vetoes of ministers as substitutive
or subverting informal powers and their outcomes 264
Table 5.5 Mapping the appointment of a minister as a substitutive
or subverting informal powers 268
Table 5.6 Informal powers in government formation per type,
outcome and country 269
Table 5.7 Mapping pocket vetos as substitutive or subverting
informal powers 273
Table 5.8 Mapping line-item vetoes as substitutive or subverting
informal powers 277
Table 5.9 Mapping line-item vetoes as parallel and risk-taking
informal powers per year 279
Table 5.10 Mapping line-item vetoes per type of informal power
and per country 284
Table 5.11 Determinants of the positive outcomes of presidential
informal activism 286
Table 6.1 Contextual factors and Personal features related
to presidential interventions 303
Table 6.2 Contextual factors and Personal features related
to presidential interventions in Germany 318
xxii LIST OF TABLES
Introduction
Presidents’ role and powers have been studied in relation to the type
of regimes they operate. In fact, the discussion about what kind of
regime type is preferable is inherently related to what kind of equilib-
rium of powers emerges between the legislative and the executive. As a
consequence, the discussion about presidential strengths or weaknesses
originates from this specific debate, of which I hereby recall the most
important steps.
Political science debate on regime types is one of the most traditional
of the entire discipline and it stems from the classic distinction between
legislative, executive and judiciary provided in Montesquieu’s theory of
the separation of powers.
However, the contemporary debate on regime types actually starts with
the article of Juan Linz (1990) related to the perils of presidentialism. In
such an article and in the following book written with Linz & Valenzuela
(1994), Linz highlights the problematic role of presidents holding many
strong powers and the consequently populist drifts. Moreover, he claims
that certain institutional features—such as the double democratic legit-
imacy and the fixed term of presidents, are likely to push presidents to
oppose congresses, especially in instances of divided government, and to
make such conflict hard to overcome.
Linz’s thesis is that parliamentarism is better and less rigid than pres-
identialism and therefore preferable especially in transitional regimes or
in fragile new-born democracies. In fact, parliamentarism is supposed to
fit in divided societies (Lijphart, 1992), as the legislative-executive conflict
can be solved with the removal of cabinet and populist drifts are less likely,
since prime ministers (PMs) are not directly elected.
Subsequently, other authors partly contested Linz and Valenzuela’s
interpretation by pointing out that presidential systems even present some
benefits (Cheibub, 2007; Morgestern et al., 2020; Shugart & Carey,
1992; Mainwaring & Shugart, 1997). For instance, presidential fixed
terms enhance government stability and continuity in public polices; and
separation of powers provides more robust checks and balances; and
ensures effective accountability as decision-makers are clearly identifi-
able. Finally, presidents are more likely to embody nationwide interests
in comparison to PMs, who are more linked to their constituencies and
parties.
When this debate, originating in the Americas where presidentialism is
the most frequent type of regime, reached Europe, where there are almost
no examples of presidential systems (with the only exception of Cyprus),
the semi-presidential regime type was put at the centre of this assessment.
In fact, semi-presidentialism is a “European institutional by-product”. On
the one hand, historical examples of semi-presidentialism can be found
in Europe since the Weimar Republic in 1919 and the Austrian First
Republic (1929). Nevertheless, examples of semi-presidential systems
were present in Europe well before the most famous prototype of the
Fifth French Republic.1 One only needs to consider cases such as the
Irish Republic (1937), the Icelandic Republic (1944) and the Finnish
Republic2 (1944). On the other hand, with the third wave of democ-
ratization (Huntington, 1991), semi-presidentialism has widely spread,
thus becoming one of the most emblematic types of regime in Europe
(i.e. Portugal since 1976 and most of Central-Eastern countries after the
collapse of the Soviet Union).
The word semi-presidentialism was first coined by Duverger (1980)
and was conceived as a sort of hybrid between parliamentarism and pres-
identialism. From the ‘90s onwards, the European debate has revolved
1 Even though presidents have been directly elected only since 1964.
2 In Finland presidents have been directly elected since 2000.
1 INTRODUCTION 3
around two main issues: (1) the problematic definition of this “new”
regime type and (2) its assessment in relation to parliamentarism and pres-
identialism. As far as the first point is concerned, Elgie (1999) pointed
out that even though many different definitions were available, they
only increased confusion rather than clarity, especially when they tried
to encompass presidential actual powers or behaviour in such classifi-
cations. As a consequence, Elgie (2011) proposed a formal definition
of semi-presidentialism which is widely accepted in literature. Elgie’s
definition include two variants of semi-presidentialism, with a clear refer-
ence to Shugart and Carey’s work (1992): premier-presidentialism and
president-parliamentarism.
The other strand of the academic debate revolves around the pros
and cons of semi-presidentialism. On the one hand, Linz and Valen-
zuela expressed for semi-presidentialism similar concerns to those related
to presidentialism (such as double democratic legitimacy, presidential
fixed terms and populist drifts) and they even added that fused power
systems increase the likelihood of executive-legislative conflicts, especially
in case of cohabitation and divided executive. On the other hand, others
(Pasquino, 1997; Sartori, 1996) sustained that semi-presidentialism is
more likely to combine the advantages of both parliamentarism and pres-
identialism and, therefore, positive elements outweigh potential pitfalls,
especially for countries in transition towards democracy. Among these
pros one can find government stability, continuity in public policy and
the possibility to replace the PM and their cabinet when the conflict
between executive and legislative arises. More recently, scholars empha-
sized that not all types of semi-presidentialism are dangerous, but only
those which establish presidents with many and strong powers, namely
president-parliamentarism cases (Elgie, 2011; Elgie & Moestrup, 2016;
Sedelius & Linde, 2018).
The short outline provided above demonstrates that presidential role
and powers have been investigated and evaluated, especially when presi-
dents operate in presidential or semi-presidential regimes. However, there
is almost no debate in political science literature about presidents in parlia-
mentary systems (see Tavits, 2009 as exception). On the one hand, this
outcome is due to the idea that presidents in parliamentary systems are
only figureheads with exclusively ceremonial roles. Consequently, it comes
with no surprise that, in parliamentary democracies, most political science
research favours investigations on PMs and their cabinets (i.e. Baylis,
1996; Helms, 2005; Jones, 1991; Rhodes & Dunleavy, 1995; Rose,
4 S. GRIMALDI
1991; Weller, 1985), since the executive power is robustly in their hands.
Even in certain types of semi-presidentialism, when presidents are gener-
ally far from day-to-day politics, political scientists’ interest in presidents’
role is somewhat limited.
In Europe, the most influential debate—partially connected with pres-
idential powers—revolves around the so-called presidentialization thesis
(Poguntke & Webb, 2005) according to which even parliamentary and
semi-presidential regimes are becoming more presidential in their actual
practice without changing their formal structure. The logic of presiden-
tialization implies the growing power and autonomy of political leaders
within political executives and political parties, and in the emergence
of increasingly leadership-centred electoral processes. To use Passarel-
li’s words (2015: 6) “the presidentialization concept combines different
empirical trends into a unique theoretical understanding of ongoing
power shifts (also) within parliamentary regimes”. The presidentializa-
tion assumption has been criticized both because of the weaknesses of its
empirical findings (Dowding, 2013; Heffernan, 2005, 2013; Karvonen,
2010) and because it underestimates how institutions clearly shape polit-
ical behaviour (Samuels & Shugart 2010). Recently, the work by Passarelli
(2015) on the presidentialization of parties confirms that the institutional
logic supported by Samuels and Shugart (2010) holds strong however,
certain political contextual dynamics, such as the parties’ genetic features,
have leverage in explaining presidentialization. On one side, this litera-
ture has been useful in pointing out that presidentialization of politics
revolves around the empowerment of the Head of Government who
can be the President or the PM especially when he/she is also the
leader of nationalized and centralized parties. On the other side, it seems
this thesis leaves little space for the Head of State with no or limited
executive role. In this respect, Passarelli’s useful distinction of presiden-
tialization, personalization and centralization of party politics helps in
understanding that presidentialization occurs in semi-presidential systems
where presidential powers are strong and significant, whereas in other
cases of semi-presidentialism or parliamentarism there is personalization
without presidentialization. Following this line of reasoning, we can argue
that the empowerment of weak Presidents (in parliamentarism and semi-
presidentialism) is always contingent and unstable since it is grounded
on personalization of politics which refers to personal capital in terms
of skills, characteristic, attitudes and reputation and not on presidential-
ization which refers to specific institutional and party resources. Finally,
1 INTRODUCTION 5
these reflections have been also relevant because they emphasized the link
between personalization and informal politics. In fact, they argue that
the empowerment of PM in parliamentary contexts “stems more from
informal practices and the circumvention of formal rules rather than from
any significant changes to the constitutional structure or to the parties
themselves” (Gauja, 2015; Passarelli, 2015: 251).
The party decay age (Mair, 2013; Manin, 1997; Tormey, 2015), which
involves most consolidated democracies, has opened new windows of
opportunity for relatively weak presidents. In fact, cases of presidents who
intervene informally in fundamental spheres of power such as government
formation and policy-making are more and more frequent in parliamen-
tary and semi-presidential systems in Western Europe and cannot be
considered odd exceptions any longer, but need to be seriously taken into
account by political scientists.
All in all, this book is an attempt to fill a gap in political science
literature by investigating the role of formally constrained presidents.
The problem with check lists is that all presidential powers have the
same weight, thus many nominee powers end up increasing the overall
score, as they are more in numbers, even though less important than
other powers (such as that of dissolving parliament). Even Shugart and
Carey’s method (1992) presents some pitfalls since it cannot evaluate
presidents’ powers irrespective of the regime type presidents operate in. In
other words, Shugart and Carey’s interest is concentrated on presidential
and semi-presidential regimes only, thus hindering their capacity to eval-
uate the powers of presidents in parliamentary systems. In fact, as Metcalf
(2000: 668) claims, they do not include presidential referral, typical
for presidents who operate in parliamentary systems, among legislative
powers, and therefore their method is not always applicable.
Other disagreements in such attempts to evaluate presidential powers
have arisen on the decision of whether to consider direct election as a
power within these check lists or not. This consideration stems from an
important debate that focuses on the correlation between direct elec-
tion and presidential activism. However, there is no clear-cut evidence to
sustain that direct election either increases (i.e. Köker, 2017; Schleiter &
Morgan-Jones, 2010) or decreases (i.e. Amorim Neto & Strøm, 2006)
presidential activity.3 It is more likely that, as finally, Tavits (2009)
clarifies, presidential activism is connected to certain opportunity frame-
works (i.e. high volatility rates; parliamentary fragmentation; minority
governments, unstable coalition governments, etc.) rather than to direct
election.
Another important problem is that among presidential formal powers,
some check lists comprise certain aspects wanting to also capture the
real praxis of the presidency (Amorin Neto & Costa Lobo, 2009;
Cranenburgh, 2008; Siarof, 2003; Tavits, 2009). However, these studies
don’t actually depart from the traditional ones, but simply re-scale the
assessment of presidential powers according to the use of certain powers
or the abstention from acting on the part of presidents. The problem
with this assessment of presidential behaviour is that it is not clear if these
scores are capturing presidential powers per se or other factors which
may have an impact on the political system such as party competition
(Doyle & Elgie, 2016: 733).
3 Especially because direct election seems to increase presidential activism when only
one type of intervention is considered, such as vetoes (Köker, 2017) or discretionary
government replacement (Schleiter & Morgan-Jones, 2009).
1 INTRODUCTION 7
As the European context is the focus of this book I can say that,
according to formal powers, weak presidents are those who score below
the mean value of 0.226 when all regimes are considered and below 0.214
when one only considers existing regimes. Based on these criteria, all pres-
idents in the parliamentary system (as expected) and some presidents who
operate in semi-presidential systems are weak, namely their score is below
the mean value as reported in Table 1.1.
Building on the idea of consistent and inconsistent constitutional
designs provided by Fruhstorfer (2019), when dealing with presidential
institutions I assess as inconsistent designs those which combine directly
elected presidents with a low score of presidential powers (namely with
a score below the mean value) or indirectly elected presidents with high
scores (namely with a score above the mean value), whereas those which
combine directly elected presidents with high scores and indirectly elected
presidents with low scores of presidential powers are defined as consistent
designs.
This exercise allows us to understand that the majority of European
presidencies (17 out of 28) seem to rely on a consistent constitutional
design whereas an important minority (11 out of 28) rely on an incon-
sistent constitutional design. On the one hand, almost all presidential
institutions in parliamentary systems are consistent in terms of constitu-
tional design, since low scores of presidential powers are associated with
indirect election. However, relevant exceptions exist such as Hungary
and Czech Republic where there are indirectly elected presidents with
relatively strong powers. On the other hand, inconsistent constitutional
designs seem to affect more semi-presidential regimes than parliamen-
tary ones, especially in Western Europe, where directly elected presidents
enjoy low powers. Among these cases, one finds Austria, Finland, France,
Ireland and Portugal.
All in all, this discussion has been useful to understand that:
(1) Regime types are not always consistent with the strength of pres-
idents. As mentioned above, some presidents in semi-presidential
systems are stronger than presidents in presidential systems; and
some presidents who operate in parliamentary system are stronger
than certain presidents in semi-presidential systems. Thus, I agree
with the claim that the internal coherence of classic categorization
of parliamentarism, semi-presidentialism and presidentialism is not
1 INTRODUCTION 9
(continued)
10 S. GRIMALDI
Notes (1) Here are reported all scores per existing regimes (namely
28 country in Europe with a directly or indirectly elected presi-
dent). However, Doyle and Elgie (2016) took into account also
previous regimes (here intended as different constitutional frame-
works where presidential powers varies) such as Albania 1991–1998;
Croatia 1991–2000; Latvia 1992–1997 etc. Therefore, 0.226 is the
mean value of 49 country time periods whereas 0.214 is the mean
value for the actual 28 regimes
(2) Russia and Turkey are geographically placed between Europe
and Asia and therefore have been excluded from this table
Source Adaptation from Doyle and Elgie (2016) and from
Fruhstorfer (2019)
Even though formal criteria adopted in literature are not fully satisfying
(because there is no doubt that formal powers cannot capture the whole
scale of presidential activism) this definition of weak presidents is suffi-
ciently rigorous to allow comparative studies and to avoid common sense
1 INTRODUCTION 11
with the only exception of French presidents (i.e. Bucur & Elgie, 2012;
Gaffney, 2012; Grossman & Sauger, 2014; Kuhn, 2010; Passarelli, 2010;
Stacey, 2013; Thiébault, 2016) who are consequently excluded in this
work. Moreover, by focusing on Western European cases where demo-
cratic institutions are consolidated, one should be able to avoid the risk of
shifting the analysis only to informal powers which produce pathological
outcomes.
This is a Small-N comparative research of most-similar cases built
according to a double-level selection. On the one hand, six country-
cases with formally weak presidencies have been selected in Western
Europe, namely with a prespow1 score below the mean value of 0.214.
More precisely, two cases (Germany and Italy) are parliamentary systems
with a consistent constitutional design; four are semi-presidential systems
with inconsistent constitutional design, and among these latter three are
premier-presidentialisms, namely Finland, Ireland and Portugal, and one
is a president-parliamentarism, namely Austria.
Further consideration should be expressed for Finland and Portugal. In
fact, according to Doyle and Elgie’s evaluation (2016) both the Finnish
and Portuguese presidencies have witnessed certain important changes
over time. In particular, the score of presidential power in Finland was
higher than the current one (0.05) in both the period 1957–1994 (0.157)
and in the period 1995–1999 (0.162). However, in both cases, these
scores are under the mean value, and therefore the president is evaluated
as formally weak for the whole time period. In Portugal, the first phase
of semi-presidentialism foresaw a relatively strong president, and from
1976–1982 the prespower1 score was above the mean value (0.264).
Therefore, President Eanes’ first term is generally not considered, with
the exception of Chapter 2, which deals with the socio-economic charac-
teristics and political career of presidents. Table 1.2 summarizes the main
characteristics for the selection of the country cases.
The country cases selected comprise a number of similarities: they
all belong to the EU, even though Germany and Italy are founding
members, whereas Ireland and Portugal have accessed later (in the mid-
70s and mid-80s, respectively) and Finland and Austria in 1995; they
are all democracies since the Second World War, with the only exception
of Portugal which has become a democracy in 1974; as far as foreign
policy is concerned, half of them have a pro-atlanticism position and are
consequently members of NATO (Germany, Italy and Portugal), whereas
the other half shares a neutral position given the peculiar history or the
24 S. GRIMALDI
Table 1.2 The main characteristics for the selection of the country cases
Source Adaptation from Doyle and Elgie (2016) and from Fruhstorfer (2019)
geopolitical position during the Cold War (Austria, Ireland and Finland).
Further, from a historical point of view, before reaching their demo-
cratic age, four countries experienced an authoritarian phase. This is the
case of Austria, Germany, Italy and Portugal, whereas two, Ireland and
Finland, were democracies well before the Second World War. In addi-
tion, four countries took part in the Second World War with a clear impact
on their polities such as Austria, Germany Italy and Finland, whereas
two remain neutral—Portugal and Ireland. Finally, two countries experi-
enced a change in their polity definition after international reconfiguration
processes: Germany, with the reunification, and Ireland with the Good
Friday Agreement. However, in most of these countries conflicts on how
to define political community and its values have often raised in connec-
tion with the bitter legacies of the past (i.e. in Austria, Germany and Italy
in relation to their Nazi or fascist past, in Ireland and Finland after the
Civil War).
When looking at their political systems, some indicators reveal certain
differences especially if adopting a longitudinal perspective. In fact, in
the post-war period and until the 1990s, Austria, Germany and Ireland
showed a low level of polarization and party fragmentation, whereas
Finland and Italy presented a higher level of polarization and party
fragmentation. From the 1990s onwards, the political environment has
changed in most countries. Electoral volatility has increased everywhere
(especially in Italy), and so has party fragmentation (especially in Finland
and Italy); polarization has increased in half of the cases, and especially in
Austria (in Italy it decreases but it is still the highest). Portugal is the only
country in countertrend, since all these indicators have decreased from
1 INTRODUCTION 25
the 90s onwards, even though this is probably due to the consolidation
of democracy and of the Portuguese party system.
The second-level selection comprises 53 presidents, namely all those
in power after the Second World War, with the exception of Irish and
Portuguese presidents which were considered from 1938 and 1975,
respectively. However, on certain occasions, this universe has been
restricted to a lower number of presidents per country, depending on data
availability. This problem has arisen specifically when studying presiden-
tial informal powers such as secondary sources or press releases as well as
presidential speeches delivered during the time frame under consideration
aren’t always available. Therefore, justifications for the selection of the
cases are provided in Chapters 4, 5 and 7 in detail. Table 1.3 shows the
share of presidents per country as well as of those related to presidential
terms.
Finally, when focusing on informal presidential interventions, the
explicative analyses is conducted by relying on 148 observations affecting
the government building and policy-making. Such informal interventions
involved 27 presidents in Western Europe.
The timeframe of this work is basically related to the establishment of
democratic regimes from the Second World War to 2021. However, there
are certain exceptions. In the case of Ireland, the whole democratic repub-
lican period which started in 1937 has been taken into account, since the
country did not take part in World War II and the time frame 1937–
1945 corresponds to the term of the first President (Douglas Hyde).
At the same time, in the case of Finnish presidents, the time frame
considered is 1944–2020, even though the democratic regime started
Table 1.3 Presidents and presidential terms per country. (Absolute and
percentage values)
Presidents N % Terms N %