how power changes hands: transition and succession in government (transforming government)

Understanding Governance seriesGeneral Editor: R.A.W. Rhodes, Professor of Government, University of Tasmaniaand Distinguished Professor of Political Science, Australian National University

Understanding Governance encompasses all theoretical approaches to the study ofgovernment and governance in advanced industrial democracies. It has three long-standing objectives:

1. To understand the process of change;2. To develop theory to explain why change occurs; and 3. To set this change and its causes in comparative perspective.

The series includes titles that adopt post-structural and post-modern approaches topolitical science and challenge such notions as hollowing-out, governance, coreexecutives, policy networks and the new institutionalism. It also publishes materialwith traditional institutional and historical approaches to such topics as primeministers, ministers, the civil service and government departments. All titles meetnot only the conventional standard of theoretical and empirical rigour but also seekto address topics of broad current interest that open the field of study to new ideasand areas of investigation.

Titles include:

Paul ’t Hart, John Uhr (editors)HOW POWER CHANGES HANDSTransition and Succession in Government

Robert Hazell, Ben Worthy and Mark GloverTHE IMPACT OF THE FREEDOM ON INFORMATION ACT ON CENTRAL

GOVERNMENT IN THE UKDoes FOI Work?

Ann ScottERNEST GOWERSPlain Words and Forgotten Deeds

Kevin TheakstonAFTER NUMBER 10Former Prime Ministers in British Politics

Titles previously published in the Transforming Government series include:

Simon Bulmer, Martin Burch, Caitríona Carter, Patricia Hogwood and Andrew ScottBRITISH DEVOLUTION AND EUROPEAN POLICY-MAKINGTransforming Britain to Multi-Level Governance

Nicholas Deakin and Richard ParryTHE TREASURY AND SOCIAL POLICYThe Contest for Control of Welfare Strategy

Neil C. M. Elder and Edward C. PageACCOUNTABILITY AND CONTROL IN NEXT STEPS AGENCIES

Oliver JamesTHE EXECUTIVE AGENCY REVOLUTION IN WHITEHALLPublic Interest Versus Bureau-Shaping Perspectives

David Marsh, David Richards and Martin J. SmithCHANGING PATTERNS OF GOVERNANCE IN THE UNITED KINGDOM Reinventing Whitehall?

Iain McLeanTHE FISCAL CRISIS OF THE UNITED KINGDOM

Edward C. Page and Vincent Wright (editors)FROM THE ACTIVE TO THE ENABLING STATEThe Changing Role of Top Officials in European Nations

Hugh PembertonPOLICY LEARNING AND BRITISH GOVERNANCE IN THE 1960s

B. Guy Peters, R. A. W. Rhodes and Vincent Wright (editors)ADMINISTERING THE SUMMITAdministration of the Core Executive in Developed Countries

R. A. W. Rhodes (editor)TRANSFORMING BRITISH GOVERNMENTVolume One: Changing InstitutionsVolume Two: Changing Roles and Relationships

David RichardsNEW LABOUR AND THE CIVIL SERVICEReconstituting the Westminster Model

Martin J. SmithTHE CORE EXECUTIVE IN BRITAIN

Kevin TheakstonLEADERSHIP IN WHITEHALL

Kevin Theakston (editor)BUREAUCRATS AND LEADERSHIP

Patrick Weller, Herman Bakvis and R. A. W. Rhodes (editors)THE HOLLOW CROWNCountervailing Trends in Core Executives

Understanding GovernanceSeries Standing Order ISBN 978-0-333-71580-2(outside North America only)

You can receive future titles in this series as they are published by placing a standing order.Please contact your bookseller or, in case of difficulty, write to us at the address below with yourname and address, the title of the series and the ISBN quoted above.

Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke,Hampshire RG21 6XS, England

How Power Changes HandsTransition and Succession in Government

Edited By

Paul ’t Hart Professor of Political Science and Professor of Public Administration, Australian NationalUniversity and Utrecht University, the Netherlands

and

John Uhr Professor of Public Policy, Crawford School of Economics and Government, AustralianNational University

Editorial matter, selection, introduction and conclusion © Paul ’t Hart andJohn Uhr 2011All remaining chapters © respective authors 2011

All rights reserved. No reproduction, copy or transmission of thispublication may be made without written permission.

No portion of this publication may be reproduced, copied or transmittedsave with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6–10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publicationmay be liable to criminal prosecution and civil claims for damages.

The authors have asserted their rights to be identified as the authors of thiswork in accordance with the Copyright, Designs and Patents Act 1988.

First published 2011 byPALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010.

Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world.

Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries

ISBN 978-0-230-24296-8 hardback

This book is printed on paper suitable for recycling and made from fullymanaged and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin.

A catalogue record for this book is available from the British Library.

A catalog record for this book is available from the Library of Congress.

10 9 8 7 6 5 4 3 2 120 19 18 17 16 15 14 13 12 11

Printed and bound in Great Britain byCPI Antony Rowe, Chippenham and Eastbourne

Contents

List of Figures vii

List of Tables viii

Acknowledgements ix

Notes on Contributors x

1 Power Transitions and Leadership Successions in Government 1Paul ’t Hart and John Uhr

Part I Understanding Transitions 21

2 Ruling Elite Transitions in Australia and the United States 23John Higley and Jan Pakulski

3 Managers or Messiahs? Prime Ministerial Leadership and the 37Transition to GovernmentJames Walter

4 Establishing Prime Ministerial Leadership Style in Opposition 55Wayne Errington

5 Bicameralism and the Dynamics of Contested Transitions 73John Uhr, Stanley Bach and Louis Massicotte

6 Westminster Norms and Caretaker Conventions: Australian 94and New Zealand Transition DebatesMarian Simms

Part II Understanding Successions 109

7 Seeking and Keeping the Hot Seat: A Comparative Analysis 111of Party Leader Successions Matthew Laing and Paul ’t Hart

8 Holding Party Leaders to Account: The Westminster Cases 133William Cross and André Blais

9 The Firing Line: When and Why Do Prime Ministers Fire 157Ministerial Colleagues?Keith Dowding and Elizabeth McLeay

10 Power Consolidation in Leadership Change Contexts: 174A Social Identity PerspectiveEmina Subasic and Katherine J. Reynolds

v

11 Building Trust at the Beginning of a New Leadership Role: 191The Role of Learning and Collaboration Paul Atkins

12 ‘The Tideless Pond that Seemed Waiting for Me’: The Afterlife 208of Australian Prime Ministers Paul Strangio

13 How Power Changes Hands: Concluding Reflections 230Paul ’t Hart and John Uhr

Index 246

vi Contents

List of Figures

7.1 Party leader longevity distribution 11310.1 Social identity dynamics of leadership consolidation 182

preceding and following transition and succession11.1 Short-term demands versus long-term interests 196

vii

List of Tables

7.1 Strategic choice for leadership aspirants and new leaders 1187.2 Distribution of aspirant/new leader types 1197.3 Succession outcomes for successor types 1197.4 A typology of incumbent leaders’ positions vis-à-vis 122

succession challenges 7.5 Primary and secondary succession triggers 1268.1 Methods by which party leaders may be removed (N = 22) 1398.2 Who can remove the leader? 1438.3 Relative security of party leaders 1478.4 Leaders formally removed (1965–2007) 148

viii

Acknowledgements

Many of the papers in this book were first presented at the Public LeadershipWorkshop, which we have hosted annually at the Australian National Uni-versity between 2007 and 2009 and which now continues to be organized byother members of the network of Australia and New Zealand based scholarsthat was thus created. We are grateful to the workshop sponsors: the AustraliaNew Zealand School of Government and its dean professor Allan Fels; theResearch School of Social Sciences and the Crawford School of Economics and Government, including the Parliamentary Studies Centre, all at the ANU.And we express our thanks to workshop participants for intellectual support,and hope that this volume will be followed by many other products of thisexciting interdisciplinary venture. As ever, Karen Tindall provided essential editorial support, as did Mark Fabian on the index.

Canberra, November 2010Paul ’t Hart and John Uhr

ix

Notes on Contributors

Editors

Paul ’t Hart is professor of political science at the Research School of SocialSciences, Australian National University, and professor of public adminis-tration, Utrecht School of Governance, Utrecht University.

John Uhr is Professor of Public Policy, Crawford School of Economics andGovernment, Australian National University.

Contributors

Paul Atkins is an associate professor at the Crawford School of Economicsand Government, Australian National University.

Stanley Bach, formerly of the Congressional Research Service, Washington, is on the panel of advisors at the Parliamentary Studies Centre, AustralianNational University.

André Blais is professor in the Department of Political Science at theUniversité de Montréal.

William Cross is the Hon. Dick and Ruth Bell Chair for the Study ofCanadian Parliamentary Democracy in the Department of Political Scienceat Carleton University.

Keith Dowding is professor of political science at the Research School ofSocial Science and director of research at the College of Arts and SocialScience, Australian National University.

Wayne Errington is lecturer in politics at the School of History andPolitics, University of Adelaide.

John Higley is professor of government and sociology at the Departmentsof Government and Sociology, University of Texas at Austin.

Matthew Laing is a doctoral candidate at the School of Politics andInternational Relations in the Research School of Social Sciences, AustralianNational University.

x

Louis Massicotte is professor in the Department of Political Science atUniversity Laval, Canada.

Elizabeth McLeay is professor in comparative politics at the School ofHistory, Philosophy, Political Science and International Relations, VictoriaUniversity of Wellington.

Jan Pakulski is professor of sociology at the School of Sociology and SocialWork, University of Tasmania.

Katherine Reynolds is an Australian Research Fellow at the Department ofPsychology, Australian National University.

Marian Simms is head of school at the School of History, Heritage andSociety, Faculty of Arts and Education, Deakin University.

Paul Strangio is a senior lecturer in Australian politics in the School ofPolitical and Social Inquiry, Monash University.

Emina Subasic is an ARC Australian Post Doctoral Fellow at the Departmentof Psychology, Australian National University.

James Walter is professor of political science in the School of Political andSocial Inquiry, Monash University.

Notes on Contributors xi

This page intentionally left blank

1Power Transitions and LeadershipSuccessions in GovernmentPaul ’t Hart and John Uhr

Power changes hands

The peaceful transition of power from one set of hands to another is one ofthe basic features of a working democracy. The simple image is of a newpair of hands replacing a former pair or indeed collection of hands, each setof hands able and willing to hold the reins of power, or in a more contem-porary image, exercise the levers of power. In some eyes, the levers arealmost neutral instruments of power which lend themselves equally to anyset of authorized hands. In democratic systems, authorization involvesdemocratic processes of election and representation, with popular support(and not mechanical competence) taken as the primary test of eligibility toexercise power. For all their many institutional complexities, democraticgovernments are designed around relationships between an evolving appa-ratus of governed instruments (‘the machinery of government’) and com-petition among potential governing hands (‘left hands’ with a bias towardshands-on government; ‘right hands’ with a bias towards hands-off govern-ment; both requiring the power of ‘many hands’ to set the levers in theirpreferred position).

Political transitions occur in every type of political regime but they areevents of high public value in democratic regimes: access to governingpower being transferred through electoral processes from one set of handsto another. Popular elections are opportunities for democratic voters topass judgement on their elected representatives and to determine whetherthe current crop of representatives should be replaced by another crop ofrepresentatives. Peaceful transitions are one very important test of a demo-cracy. Regimes of representative democracy operate through regular elec-tions which frequently give rise not only to government transitions but also to office-holder successions in the ranks of government and furtherdown. With every change in the electorate’s choice, there are transitions in and out of elected office which in turn trigger significant successions and turnovers in the unelected offices (bureaucratic, military, judicial)

1

established under the control of government to carry out important publicservices. These movements in and out of government are driven by thelogic of the electoral process, which holds that those holding the powers of government get their democratic authority from the electorate. Thus,democracies are sustained by a commitment to electoral processes whichthemselves allow for variety and change as voters register their evolvingpreferences for those individuals and political parties which should nexthold the powers of government.

In this volume, we approach ‘transitions’ generally in terms of the arrival of an alternative set of hands ready to replace the existing team and to exer-cise the levers of power. So too we think of ‘successions’ as successive replace-ments of individual power-holders by others, which can be part of suchtransitions but can also comprise individual retirements, sackings or within-government portfolio reshuffles, mergers and splits. We can begin to think oftransitions as typically emerging from interparty competition (oppositionsreplacing governing parties or coalitions) and of successions as emerging fromintraparty competition (e.g. factional rivalries and leadership struggles). Bothtypes of change are innovations, even though both can be driven by quiteconservative political and social commitments. Both types of innovationchange the personnel of leaders exercising the relevant powers and both havethe potential to change the style of leadership exercised by ‘the new team’. In this volume, we study their antecedents, dynamics and effects.

Turnover at the top: Transitions and successions

Transitions: Characteristics and challenges

In this volume, we use the term transition to denote a change in the partyor coalition that holds government. Transitions imply that multiple ministersor political appointees are rotating at the same time, and that all or many ofthe newcomers experience a shift from campaigning, waging opposition orholding non-political offices to running executive government. Transitionsimply that new political bargains about the content and priorities of gov-ernment policy are being struck. They entail a period of uncertainty and dis-continuity for the swathe of departments, agencies and stakeholders involvedin shaping and implementing public policy, as well as for the media and thegeneral public.

In democracies transitions are usually triggered by elections. In par-liamentary democracies with majoritarian electoral systems, the electionresult tends to be decisive in determining who will govern and who will bein opposition in the next parliament. The new prime minister is invariablythe leader of the party with the largest number of seats in the lower house.New governments can be sworn in quickly, as portfolio allocation tends tobe a relatively straightforward, mostly intraparty affair. In systems withproportional representation, the number of parties in parliament tends to

2 Power Transitions and Leadership Successions in Government

be larger. No single party achieves a parliamentary majority and coalitionsbetween parties tend to be more fluid and opportunistic. Often, intense,sometimes protracted post-election coalition negotiations are necessary,and outcomes are not always easy to predict. In presidential systems, thereis effectively a situation of divided government, in which one party mayend up holding the presidency (and thus executive government) but notcontrolling a parliamentary majority, and vice versa. To some extent there-fore, dual (executive and parliamentary) or even triple (if lower and upperhouse both see shifting majorities) power shifts may result from elections.However in this volume we will use the term transition exclusively forchanges in party control of executive government.

Though by far the most prevalent, elections are not the only mechanismby which transitions may come about. Extraordinary circumstances such aswar may bring about shifts towards a more inclusive ‘national government’transcending the usual government-opposition divide. This was the case inmany countries after the outbreak of World War II. Conversely, when rela-tions within the governing coalition suffer a break down in mid-term,reconfigurations may occur. Parties may withdraw from the coalition whileothers continue (sometimes as a minority government) or co-opt new part-ners among the hitherto non-governing parties to join into a new coalition.Voters may not be given a say in such cases until much later. Examples of unstable coalition governments experiencing frequent, non-electorallyinduced transitions have abounded in post-war Italian, Belgian and, to alesser extent, Dutch politics.

Finally, in cases of exceptionally paralytic standoffs between the exec-utive and the parliament, in some countries the head of state might make atie-breaking decision that implies a standing down of the incumbent headof government and their party/coalition, and the imposition of some formof caretaker regime paving the way for new elections. Such situations are by definition extremely rare, and bound to be deeply controversial (as inthe case of the sacking by the Governor-General of the Australian LaborParty government led by Gough Whitlam in November 1975). In semi-presidential systems in particular, dissolutions of parliament by the head ofstate can be part and parcel of power struggles between the head of state and the parliamentary majority induced by constitutional ambiguity concern-ing the scope and limits of their respective responsibilities (Elgie, 1999). As in all these cases significant changes in party control of executive governmenteventually occur, they are all encompassed in our definition of transitions.

The term ‘democratic transitions’ usually refers to the historical processof regime change managed by non-democratic regimes moving along thepath towards a more openly democratic future. South Africa is a good recentexample involving that titan of transition, Nelson Mandela, bringing funda-mental change to a traditional political system with a history of minoritywhite rule over an increasingly-unwilling non-white majority. Democratic

Paul ’t Hart and John Uhr 3

transitions are tested by their degree of sustainability, which attracts therelated term ‘democratic consolidation’: the effectiveness of such regimetransitions is measured not simply by the appointment of a Mandela figureto the top political office, but by the pace and extent of ‘consolidation’ ofvarious democratic attributes, such as regular elections, subsequent changesof government, rulers’ respect for the separation of powers, officials’ respectfor limited government, the community’s respect for the rule of law, andthe slow and steady extension of dispersed public powers in the context ofconstitutional checks and balances.

Does ‘power’ really ‘change hands’ when a change of government (i.e. atransition) takes place? There is no straightforward answer to this question.For some, all office-seeking parties in a democracy are part and parcel of the same centrist ‘power elite’, or, more insidiously, in fact peripheral to that elite no matter which political offices they hold. For others, the checks placedon the executive power of political office-holders have become so strong andthe relative autonomy of the state’s core institutions so limited that ‘whogoverns’ is no longer the million dollar question. In a world where power isactively dispersed as well as becoming fragmented, we should instead wonder‘where and how governance happens’ (Rhodes, 1997). This suggests that thereis merit in remaining mindful of the distinction between two related tran-sitions: transiting into the forms of public office and into the substance of real power. Many reformist political parties find that ‘the system’ has ways offiltering out discordant elements: e.g. by relying on systems of checks and balances to allow a reformist party to ‘take office’ but to prevent it from really‘taking power’. The roots are embedded in liberal constitutionalism, whichconstrains and limits the power of government. Frustration in transitions frequently arises from the carefully arranged dispersals of power found inmany democratic settings.

Democracies vary in degrees of popular control over power-holders, withpublic power itself usually divided into separate branches of governmentexercising legislative, executive and judicial power. Voters in democraticsystems have quite different degrees of popular control over power-holdersin each branch of government; indeed, even with particular branches suchas the executive branch, voters have significant say over who holds powerin the legislature but usually little say over who holds administrative powerswithin the civil or military service or in the judiciary (Kane et al., 2009).

Liberal democratic forms of governance include many checks and balancesdesigned to limit the reach of holders of executive power. The distinctionbetween winning office and winning power suggests the persistence of manysuch mechanisms, of which entrenched bureaucratic power is simply one. Buteven in democratic regimes, a minor miracle occurs in that the civilian armyof state employees by and large accepts its place as an instrument of publicadministration regulated under law by elected representatives whose leadinghands enlist, and are not enlisted by, the administrative arm of government.


Building on more than a century of public law, modern democracieshonour the values of administrative professionalism and political impar-tiality that complement the democratic norms of popular control andresponsiveness. The result is that ‘the merit system’ defines a wide realm of largely self-regulating autonomy that is surprisingly free of the sur-face friction of partisan politics. But bureaucracies, of course, have theirown endemic forms of partisanship which feature in some of the chaptersin this volume.

Successions: Characteristics and challenges

In contrast, we use the term succession to denote personnel changes inindividual leadership positions within government (or political parties).The most dramatic succession of all is that between holders of the head ofgovernment position. However, for the departments and sectors concerned,individual ministerial successions can also be highly consequential events,even when they are the product of ad hoc resignations or cabinet reshuffleswithin one and the same government.

Likewise, leadership successions in political parties and government agen-cies are high-stakes processes that may affect the politics and policies ofincumbent governments. Each transition entails a number of successions, butnot every succession necessarily comes about as part of a transition. Individualoffice-holders can die, fall ill, be fired, be promoted, retire, lose an election, orotherwise be induced to change roles. Successions can be planned, as in caseof fixed-term CEO contracts or political term limits. They can be predictableand protracted, as in the case of a minister, head of government or partyleader whose political capital is in exorable decline as a result of repeatedpolitical defeats or policy failures, or highly controversial personal foibles(such as extramarital affairs or lavish use of public funds for not-so-public pur-poses). Successions can also become necessary in a totally surprising and evenshocking fashion, for example when heads of government die in office, suchas Australian prime minister Harold Holt, who drowned during recreationalswimming in 1966, and Polish president Kaczynski, who perished in a 2010plane crash along with numerous other senior members of the Polish govern-ment establishment.

In the case of planned and predictable successions, the anticipation processcreates a whole dynamic of growing ‘lame duckness’ for the incumbent, strug-gles to counter the trend and possibly neutralize competitors, possibleattempts by the incumbent or other senior figures to groom a particular suc-cessor, and intensive lobbying in the lead-up to formal replacement decisions.When vacancies arise unexpectedly, improvisation is the name of the game,and a sense of shock and time pressure may render the successor selectionprocess that follows both emotionally intense and politically risky.

Political successions of whatever kind will inevitably elicit considerablepublic speculation. At the same time, they often feature intensive behind-


the-scenes canvassing of and manoeuvring by succession candidates and factions within the party and/or government. They are periods of highpolitical drama, and many of them involve significant doses of inter-personal, intraparty, and – when applicable – intragovernmental conflict.They may also set in motion chain reactions, with one succession trig-gering a series of people moving roles. Overall, they create periods of instability at the top, which can be traumatic and damaging, but whichalso present opportunities for reshaping the party’s or the government’spriorities and image.

Public service successions are ostensibly more orderly. They are at leastformally governed by Weberian-style procedures of merit-based selection,recruitment and performance assessment we have come to associate with aprofessional bureaucracy. In reality, the careers of top-level bureaucrats are in no small measure influenced by ‘non-professional’ political factors,of which there can be many: general pressures to arrive at a more‘representative’ bureaucracy (in terms of gender, age, ethnicity); ministerialdesires to surround themselves with ‘people they can work with’ (codefor: people who have a more acute ‘political antenna’, are more pliable, or are card-carrying party members); a post-transition climate of mistrust of the bureaucracy by the incoming government. The need to build andmaintain the thrust of the government and the minister of the day willalways be on the minds of those who select and appoint senior bureaucrats.Who the latter are (and who not) in and of itself constitutes an importantindicator of the rules of the game governing the nexus between politics(‘democracy’) and the public service (‘bureaucracy’).

Both types of ‘power changing hands’ therefore merit investigation intheir own right, though they partially overlap. This volume is the first everattempt to juxtapose both forms of ‘power changing hands’ – transitionsand successions – and explore their antecedents and implications in onestudy. The scope of this book is limited to transitions and successionwithin democratic executive governments and the political parties thatsustain them. We do not claim that its insights apply to transitions andsuccessions in other segments of society. Moreover, the volume does notengage in other pivotal forms of ‘transitology’, namely studying changes in the underlying political regime to and from ‘democracy’ (e.g. Linz and Stepan, 1978, 1996; Offe, 1996). Nor does it look at switches betweendifferent types of electoral system (e.g. the 1990s New Zealand change from majoritarian to proportional representation electoral arrangements).Finally, it also does not focus on initiatives to replace one set of constitu-tional arrangements by another (e.g. from monarchy to republic, as unsuc-cessfully attempted in Australia; from unitary or federal to confederalpolitical structures, as in Belgium and possibly Canada; or the balancebetween the prerogatives of national and supranational bodies, as in theEuropean Union).


Studying transitions and successions: Methodological considerations

Some distinctive methodological challenges present themselves to thosewho, like us, want to study transitions and successions empirically. First,though most of the chapters in this volume explicitly embrace an actor-centred approach to analysing their subject, there is no getting away fromthe realities of structure-driven forces. Like almost any other set of socialscientists, students of transition and succession need to ask themselveswhich contextual factors shape the considerations and choices of actors inthese dramas, as well as, how and to what extent (Goodin and Tilly, 2006).What role, for example, do institutional variables, such as electoral arrange-ments and party systems play in shaping the course and outcomes of tran-sitions? To what extent do actors feel constrained or encouraged byconstitutional or party ‘traditions’? And to what extent do they actively re-interpret and strive to alter the meaning attached to these traditions to suittheir present purposes in, for example, ongoing leadership struggles(Rhodes et al., 2009)? The more useful studies of transition and successionare likely to be explicitly comparative, and factor in these questions insteadof organizing them out.

Second, do we see transitions and successions as discrete events or asunfolding processes? The former perspective facilitates straightforward andnumerical description of key parameters of these events, their triggers andtheir outcomes. It allows for large-N studies. The latter encourages a morein-depth mode of analysis that stays closer to the perceptions and decisionsof key actors and picks up on the dramatic twists and turns that are partand parcel of many transition and succession episodes. Its requirementslimit the scope for large-N study, and conduce towards the kind of smaller-N ‘focused comparison’ recommended by George and Bennett (2005) andother proponents of case-study focused approaches to social science research.This volume features examples of both approaches, each with their own setof strengths and limitations.

Thirdly, notwithstanding these differences, all attempts to analyse tran-sitions and successions face the additional challenge of demarcation. Whendoes a transition or succession episode begin and end? In the era of the‘permanent campaign’, the distinction between campaigning and govern-ing is said to have lost much of its meaning, and it follows that it may benaïve to assume that the anticipation of potential transitions begins onlywhen an election is formally called (Ornstein and Mann, 2000). Likewise,in the case of succession, the eventual resignation and replacement of anoffice-holder is often the product of a long gestation period, whose originsare not easy to determine unequivocally. The sample applies to the backend of transitions and successions. When do new governments and newoffice-holders stop being ‘new’ and become ‘settled’ incumbents? What


time lines for institutionalization of new actors and regimes are sensible – and fair – to maintain in assessing their ‘performance’ and the outcomesof transitory processes?

There is no natural Archimedean point here. In assessing transitions,much is made of the ‘100 days’ mark, but this is based on the by nowalmost mythical example of the initial Roosevelt administration, whichtook office at a time of acute crisis (Alter, 2007). Likewise, the oft-used one year mark for assessing new governments and new leaders is arbitrary.Governments and their leaders in fact labour under electoral cycles of different amplitudes (e.g. three years in New Zealand and Australia; five years for French presidents; four years in most parliamentary demo-cracies), which no doubt colour the expectations concerning the consolid-ation of new rulers. Within the public service, the terms of employment of senior executives have widely different time horizons (roughly fromthree years to life tenure) between and even within jurisdictions, again sug-gesting different time scales against which their ‘settling in’ is beingassessed. These demarcation problems cannot be solved at the stroke of apen. All we can expect is that students of transitions and successions clarifyand justify the temporal frames they use when making these judgementcalls.

A final methodological challenge is that of dealing with so-called ‘non-events’. This is particularly acute in the analysis of leadership succession.There are many instances in which an ambitious backbencher or ministersconsiders challenging the incumbent leader but in the end chooses not todo the deed. How do analysts standing outside the inner circles of the partyor cabinet know when these cases have occurred? Much of the informalsoundings that go on are meant to remain secret, and in all likelihoodquite a few of them do for quite some time. And yet for those in the know,they are meaningful occurrences, whose results become part of the prismwith which they assess incumbents and their potential challengers. In thatsense, even totally informal and ostensibly unsuccessful consultationsabout the leadership can serve to destabilize incumbents and further dividecabinets and parties. At the very least, they help raise the hitherto perhapstaboo issue of the ‘sell by’ date of the incumbent. Indeed, one might arguethat such aborted challenges act as precursors to the eventual fall of theincumbent, perhaps at the hand of another challenger. Generalizationsalong these lines are difficult to sustain, however, counterexamples are nothard to come by: think of the ability of Australian prime minister JohnHoward to consistently stare down his leadership rival within the LiberalParty, treasurer Peter Costello, who never formally challenged despitegrowing encouragement from his supporters to do so during the final yearof the long-term government. (Howard eventually lost his seat and there-fore his leadership in the 2007 election, during which Labour gainedpower).


In some parties, there are also numerous unsuccessful challenges, wherethe aspirant does ‘come out’ and the matter is put to a formal vote, but isrepudiated by the relevant selectorate. Are these to be simply written off asnon-events, or are they in fact consequential – even necessary – steps in thedestruction of the political capital of an incumbent? One needs to knowmuch about the political context of the moment and calculations of theactors involved to be able to answer that question sensibly. Many chal-lenges are in fact exercises in ‘testing the waters’. No one expects them tosucceed in toppling the incumbent there and then. Their results allow par-ticipants and observers to take stock of potential shifts in the balance offorces within the party/government. Generally speaking, one may expectthat the slimmer the majority obtained by the incumbent in a challenge,the higher the likelihood of another challenge arising relatively quickly,and the higher the likelihood that the incumbent will eventually be forcedout of office (either by a future vote or by informal resignation pressures).But it is hard to test such propositions empirically as the results of partyroom voting are not always made public.

Both types of ‘non-events’ impose limits on the ability of traditionallarge-N dataset studies to fully capture the dynamics of leadership suc-cession. They only emerge on analysts’ radars when they really drill downinto the inner life of the party/government at the time. So whilst not deny-ing the considerable virtues and practical possibilities of large-N studies inthis field, ideally, studies of leadership succession combine large-N, events-focused approaches with small-N, context-sensitive and process-focusedapproaches in their overall design.

Understanding transitions: Review and agenda

One reason for this volume is that there is surprisingly little systematicresearch of the dynamics of power transitions associated with the arrivalsand departures of new and old governments in contemporary politicalsystems. There are a few important exceptions. One the one hand, there isa large literature in the field of international relations on the theme ofpower transitions, where this is usually an investigation of the changing‘balance of power’ relationships among nation states. This literatureincludes careful attention to the comings and goings of ruling elites inglobally important but non-democratic nation states. Thus, the pathologyof power transitions in authoritarian states does receive significant researchattention. Indeed, so far probably greater attention has been paid to power transitions in non-democratic systems than in democratic systems.One important reason might be the belief among researchers that demo-cratic transitions are comparatively well ‘institutionalized’ compared towhat appear to be more ‘personalized’ processes of power transition in non-democratic systems. Hence, researchers examining transitions in


non-democratic systems have relatively few predicators of who will emergeon top in the next phase of transition or succession. Compare this situationwith those researchers examining democratic systems who can confidentlypredict that the next phases of transition will bring forward the leadershipteam dominating those political parties likely to win either public confid-ence or the confidence of the representative assembly with power to deter-mine who rules. Democratic power transitions sound like a simple story ofwell-scripted process management; but such a view is a gross simplification.The general script may be there, but as noted the mixture of power, per-sonalities, institutions and contexts can and does produce a significant andnon-trivial variety of transition dramas.

There is also a large literature in the field of democratization studieswhich investigates power transitions associated with the sequencing ofregime changes from ‘pre’ to ‘proto’ to ‘consolidated’ democracies. What is missing is any attempt to stock-take the forms of power transition thatoccur within existing democratic systems. To be sure, there is an extensiveempirical literature chronicling the specifics of individual election struggles(particularly allegedly regime-shifting ‘realignment elections’), cabinet formation processes, as well as dramatic non-electorally induced govern-ment transition processes across contemporary democracies (e.g. constitu-tional crises). But what has been missing is an attempt to interpret thesevarious modes and moments of transition from a broader comparative per-spective. This collection is designed to begin to generate useful hypothesessuitable for subsequent comparative investigation across the many concreteinstances of the arrivals and departures of governing teams.

We have suggested above that a number of dimensions of the transitionsprocess are important ingredients in the explanation of success and failurein power transitions: the front-stage dimension of the changes in the handsholding the reins of executive power; the back-stage dimension of thechanges in political development that bring forward a system of democraticgovernance out of a pre-democratic system; and the off-stage dimension of the challenges posed by non-executive powers (legislative, judicial or indeed executive-bureaucratic or executive-military) to incoming gov-ernments. We now suggest that at least three analytical perspectives orsocial science orientations deserve a place in the mix of analysis of powertransitions.

First, a constitutional perspective that attempts to make sense of the designprinciples evident in the various ways that ‘the rules of the game’ of demo-cratic governance shape power transitions. The large picture emerges in thechapter by Higley and Pakulski, which brings to the fore the constitutionalnorms favoured by different types of governing elites. Adapting traditionaltypologies of ‘lions’ and foxes’, the authors review the US and Australianexperience, common to the larger ‘Anglosphere’, of alternating governingnorms associated with these two ruling types: lions tending to reshape the


rules of the game to favour their clear and open domination of democraticprocesses, and foxes tending to working around the legal limits and insinu-ate themselves into the inherited system, without drawing unnecessarypublic attention to their institutional innovations.

The chapter by Uhr, Bach and Massicotte examines three federal casesstudies, noting the various ways that bicameralism can modify transitionsplans. This chapter contrasts conventional accounts of power transitionwhich focus on the changing hands managing the reins of executive powerwith an often-neglected focus on who controls of the reins of legislativepower. Federal constitutional arrangements provide important test casesallowing researchers to relate two branches of transition: the executivebranch story which highlights the centre-stage story of new arrivals and the legislative branch story which highlights the off-stage story of non-departing challengers to the new governing team.

The chapter by Simms explores a related aspect of constitutional formsand due processes: the rules of the game known as ‘caretaker conventions’designed to regulate the ways that serving governments manage pre-election tensions, including the temptations to use the power of incum-bency to spoil the chances of challengers. Fascinatingly, these importantrules of parliamentary politics are typically devised by incumbent govern-ments advertising their commitment to free and fair transitions in theevent that their challengers might win the coming parliamentary election.The New Zealand case studies that Simms uses detail many of the ways thatthe parliamentary rules of pre-electoral contests impact on the process andsubstance of power transitions.

Second, a theory of transition leadership reflecting on the human factors atwork in striking a particular balance between stability and change in bothagendas and styles of governing. Opposition leaders who win governmentthrough parliamentary elections have choices about how to prepare for thetransition game. They may position themselves as saviours of the regimevalues that incumbents have allegedly ruined, as Walter suggests in hischapter about prime ministerial modes of transition management. Theymay also present themselves as ‘safe alternatives’ to successful but nowjaded incumbents, as Errington notes in his comparative analysis of therole of oppositional experience in shaping the leadership style of incomingprime ministers.

For some political systems, certain types of power transitions are notacceptable: consider, for example, the strong social preference among manyconsociational democracies for change that is authorized by ‘buy-in’ acrossa broad range of political elites, which can be contrasted with the syn-drome of wholesale change in ruling parties favoured by adversarial politicalregimes found in Westminster-derived systems, with clear lines separatingpolitical winners from political losers. Using contrasting Australian andBritish case materials both Walter and Errington show that it is one thing


to win office and quite another thing to transit into effective public man-agement of governmental processes and to acquire a truly ‘national’ leader-ship legitimacy. No golden formula exists, as personal styles and skillsmodify the extent to which new leaders are willing and able to abide byinstitutional transition scripts.

Subasic and Reynolds’ chapter offers a theoretical perspective on thequestion how new leaders consolidate power and authority (following atransition or a succession). Their work is rooted in social psychology, par-ticularly the influential paradigm of social identity theory. They emphasizethe importance of ‘identity entrepreneurship’, of new leaders attempting to(re)define the relevant boundaries and norms of the community (the ‘we’)they are proposing to lead. By doing so, they also create a ‘not-we’ (out-groups), and a crucial point for analysis is how these outgroups get definedand treated in the process of consolidating the bond between themselvesand the ‘we’ that leaders attempt to forge in order to legitimize and con-solidate their leadership claim. Subasic and Reynolds present a model ofleadership consolidation in which this leveraging solidarity and division in relation to ingroups and outgroups prior to and following transitions orsuccessions becomes the crucial mechanism.

Finally, there is the executive government perspective. This helps us tounderstand why some power transitions succeed and others fail by refer-ring to the managerial capacity of ‘the system’ of bureaucratic professionalsto give effect to the policy initiatives of incoming governments. Instead ithighlights the strengths but also the limitations of state structures for facil-itating power transitions. Both should be taken into consideration. Thebureaucracy in contemporary governments is not only big and all encom-passing, it is also trying hard to strike a balance between democraticresponsiveness and professional administrative autonomy. In times of tran-sition, the new rulers are most keen to see evidence of the former, and lesskeen on the latter. Since none can really do without the other in makingexecutive government work successfully, the challenge becomes one offlexibility, trust-building and renegotiating the ‘public service bargain’(Hood and Lodge, 2006).

Understanding successions: Review and agenda

As noted above, succession research focuses on the coming and going ofindividual office-holders, whether they be party leaders/prime ministers,cabinet ministers, or senior public servants. It is about the factors thatdetermine the political and professional fates of individuals, not of entiregovernments. That said, succession research is motivated in large part bythe assertion that who holds a particular high public office matters, not justfor that individual per se, but for the party that put them there, the organ-ization that is being led by them or the direction of public policy in the


relevant portfolio. Succession research differs from the long-standing socio-logical interest in the patterns of elite recruitment in and around theorgans of the state (e.g. Mills, 1956; Miliband, 1973; Borchert and Zeiss,2003; Dogan, 2003). It looks at the structures and processes by which indi-viduals move, or are being moved, into and out of these positions. To whatextent can they control their own fates? What is the professional life expect-ancy of various types of office-holders in different political systems, andhow can similarities and differences in that life expectancy be explained andassessed? What patterns of patronage, responsiveness and accountability areestablished by alternative mechanisms of selecting and removing political-administrative leaders? These are some of the key questions of comparativesuccession analysis.

There is a modest amount of prior research in this area. Most of it deals withparty leader and prime ministerial succession. Much of the work pertains to descriptive studies of individual countries or parties (e.g. Jackson, 1975;Colome and Lopez Nieto, 1993; De Winter, 1993; Marsh, 1993; Strøm, 1993;Courtney, 1995; Bille, 1997; Bynander and ’t Hart, 2007; Kenig, 2009). Thereare also plenty of journalistic or historians’ in-depth accounts of particularlygripping succession struggles (Reid, 1969; Alexandre, 1970; Baring, 1982;Kelly, 1984; Koerfer, 1987; Naughtie, 2002; ’t Hart, 2007). However, when itcomes to systematic comparative research into leadership succession in liberaldemocratic parties and governments, the numbers of studies dwindle fast (see,e.g. Calvert, 1987b; Davis, 1998; Hermann, 2005; Bynander and ’t Hart, 2006;Andrews and Jackman, 2008). Studies explicitly concerned with the specificissues of prime ministerial successions (Weller, 1983, 1985: 45–71, 1994;Alderman and Smith, 1990) are especially few and far between. The bulk ofthe existing succession research pertains to communist and autocratic regimeswhere drastically different rules of the game apply, or deals with chief exec-utives in presidential systems (McCauley and Carter, 1986; Calvert, 1987a;Swaine, 1992; Wong and Yongnian, 2002; Bueno de Mesquita et al., 2003).

Of the parliamentary democracies, the UK case is well written up, almostredundantly so (e.g. Alderman and Smith, 1990; Punnett, 1992; Stark,1996; Alderman, 1999; Quinn, 2004, 2005; Denham and O’Hara, 2008;Heppell, 2008, 2010; Denham, 2009; and on ministerial turnover Dowdingand Kang, 1998; Berlinski et al., 2007, 2010). The more analytical compar-ative contributions mostly examine successions from the perspective ofpolitical parties. In particular they focus on examining the impact of thedifferent rules and mechanisms that various parties have adopted to selectand remove their leaders. Many established parties around the democraticworld have made changes to those rules in recent times, usually favouringthe involvement of ordinary members in leadership choices, providing aseries of the kind of natural experiments that social scientists love. There is,consequently a lively debate on the origins and political dynamics of suchreforms, as well as preliminary assessments on their pros and cons (LeDuc,


2001; Pennings and Hazan, 2001; Rahat and Hazan, 2001; Bueno de Mesquitaet al., 2003; Quinn, 2004, 2005; Bynander and ’t Hart, 2007; Kenig, 2008).

More recently, a resurgence of interest in the career dynamics of cabinetministers is evident (Dewan and Dowding, 2005; Berlinksi et al., 2007;Dowding and Dumont, 2008; cf. Blondel, 1982). The ups and downs ofministerial tenures are not just described in their own right, they are alsostudied in relation to the power and leadership style of prime ministerswho, to various degrees in different polities, hire and fire prime ministers as part of their own ongoing efforts to increase and retain their politicalcapital.

What about succession of public sector CEOs and other senior public ser-vants themselves? There is a copious prescriptive literature on succession‘planning’ and ‘management,’ much of it formulaic and repetitive and notgrounded in empirical research on leadership succession in organizations, of which there is some within in the field of public sector Human ResourcesManagement (Perry and Hondeghem, 2008; Raffel et al., 2009) and a great deal within business studies (overviews in Kesner and Sebora, 1994;Giambatista et al., 2005).

In all, there is not yet a body of competing theories, or even few empir-ically well-established individual propositions entrenched on political andpublic sector succession either as a cause or as a consequence of other polit-ical phenomena (yet see Bunce, 1981). The relatively underdeveloped stateof theory development and empirical knowledge about public leadershipsuccession is somewhat surprising given the enormous amount of attentiondevoted to leadership succession speculation and interpretation in the massmedia. Political and CEO successions enjoy a prominent place in publicservice folklore where the coming and going of ministers, political appointeesand chief executives constitutes a recurrent period of uncertainty and change.Each change of masters constitutes a test of public service adaptability. Andeach is potentially consequential, as the differences of priorities, style, com-petence and clout between the outgoing and incoming office-holders maybear upon the agency’s place in interdepartmental, and indeed cabinet, peck-ing orders. The CEO succession literature in business studies is crystal clear:CEO succession matters – perceptually, psychologically and, often at least,materially (Huson et al., 2004; Rowe et al., 2005; Ndofor et al., 2009).

Given the absence of a strong pre-existing body of knowledge, whatmight be some productive lines of inquiry for political science students ofsuccession in democratic governments? We mention three here, which weshall return to in the final chapter, which compares and integrates thefindings of all the contributions to this volume. Firstly, from a power per-spective, the lens can be turned on the question of who controls whom. Itsees the rise, tenure and departure of party leaders in terms of the ongoingpulling and hauling not just between individual members of the partyand/or bureaucratic elite but between different offices and constituencies


within the party organization and executive government. For example, it understands ministerial turnover in terms of the relative power of theprime minister vis-à-vis their cabinet colleagues, backbenchers and partyforums. The debates about the rise of prime ministerial government (of the1970s and 80s) and ‘presidentialization’ of prime ministers (of the mid-1990s to the present day) are driven by considerations of power and authority.The degree to which prime ministers possess the ability to ‘hire and fire’ministers more or less at will is taken to be one important indicator ofprime ministerial dominance in these debates. More generally, a power per-spective on political succession invites us to analyse successions asprocesses of strategic actors choices, as well signalling, bargaining and evencrisis management, including the role that cognitive, emotional and social-psychological factors may play in shaping the course and outcomes ofthese processes (Bynander and ’t Hart, 2006, 2008).

Likewise, a power perspective interprets the origins and effects of min-isterial and senior agency executive turnovers within cabinet portfolios in terms of the balance of influence in political-administrative relationswithin the relevant agency. Again the question of the ability to ‘hire andfire’ is a focal point for analysis and debate. For example, questions includethe following: What does the move from permanent to fixed-term con-tracts for agency heads that we have seen in many countries imply for thepolitical control of the bureaucracy, and for bureaucratic responsiveness tothe members of the government of the day? What are the implications ofdressing up or down the role of independent public service commissionersin the appointment and assessment of senior executives? And who controlsthe terms of the performance agreements that govern the relationshipsbetween ministers and their chief executives? (Hood and Peters, 1994;Hood and Lodge, 2006). Normatively, scholars and practitioners have debatedthe relative merits of recent or planned changes with regards to each ofthese pivotal governance mechanisms at the top of public agencies (Savoie,2003; Mulgan, 2007, 2008; Podger, 2007a, 2007b; Shergold, 2007).

Secondly, the flip side of the power coin is that of accountability. Anaccountability perspective on succession generates a set of related, but analytic-ally distinct set of questions (Mulgan, 2003; Uhr, 2005). Which mecha-nisms are in place to ensure that holders of high office, who are put thereto act as ‘agents’ on behalf of some constituency and/or set of values andinterests, are induced to render account of their behaviour to the ‘prin-cipals’ who put them there? Who are in effect the relevant principals for, say,party leaders, cabinet ministers or departmental secretaries, and how are theaccountability relationships between principals and agents constituted?

The many instances of change to party rules of leadership selection andremoval that we have seen across the democratic world in the last threedecades have in large part been motivated by the idea of opening up thesepivotal leadership processes up to broader scrutiny and indeed participation,


even down to the level of ordinary party members. A key question gener-ated by the accountability perspective is to what extent these expectationshave materialized, and what their implications are for the nature of theprincipal-agent relationships between citizen-members, party apparatchiks,members of parliament, ministers and indeed heads of government.

This leads into a third area of interest, which is that of evaluation. Howdo we know a ‘good’, ‘well-managed’ succession if we see it? As does anyother form of evaluation in the world of government, assessing elite suc-cessions involves tricky issues of criteria choice, measurement and causalattribution. There are so many potential vantage points: government stabil-ity, elite accountability, legislative control of the executive, democratiza-tion of party organizations, party electability, and so on. To our knowledge,political scientists have so far eschewed this challenge: there is no normativetheory of succession. There are dilemmas aplenty. How, for example, do wedesign succession practices that navigate the tension between the need fora certain measure of stability and predictability of government behaviour,with the need for responsiveness to electoral signals, new leaders’ mandatesand indeed the need for periodic ‘creative destruction’ and course changes?

In the absence of existing frameworks, let us propose a provisional set ofcriteria for evaluating successions. If only to encourage readers to keep inmind issues of evaluation and possible redesign of succession arrangementswhen digesting the studies in Part II of this volume. We shall revisit it inthe concluding chapter, in light of what we will know by then. In our mind,an effective leader succession process meets the following criteria:

– key decisions about leader selection and removal are taken in a transpar-ent, inclusiveness and undisputed fashion;

– it produces new office-holders the party, or agency/sector can live withfor a while;

– it allows the party and/or the government (or individual agencies withinit) to consolidate as well as adapt in the face of contextual changes;

– it noticeably improves party or agency performance on its performancecriteria of choice.

The second part of the volume is devoted to comparative studies of suc-cessions in and around government. Dowding and McLeay kick off theseries by comparing UK and New Zealand prime minister’s institutionalpossibilities and personal styles when it comes to hiring, moving and firingcabinet ministers. Their examination throws up considerable institutionaldifferences between the two countries, with New Zealand prime ministersin particular needing to constantly watch their backs in relation to theircabinet colleagues, given the ease with which they can be removed by theirparliamentary parties and the relentlessness of the electoral pressures gen-erated by a three year cycle. Taking these and the personal idiosyncrasies of


the post-WWII office-holders in both countries into account, Dowding andMcLeay present us with a four-fold typology of prime ministerial styles of managing (and firing) their ministers that forms a useful starting pointfor further inquiry.

The next two chapters both focus on the succession of party leaders(whether in or out of government office). Focusing on the key Westminsterdemocracies, Cross and Blais’ (2009) large-N study examines when and howparties can hold their leaders accountable, or, in more robust parlance, howthey can get rid of leaders they no longer want. Studying both formal rulesand de facto practices, they note important cross-party and cross-nationaldifferences in the way parties handle the politics of leader ‘deselection’ anddiscuss their implications.

Laing and ’t Hart present findings from a new dataset comprising 515succession cases covering 66 political parties in 23 established democraciesin the post-World War II period. They specifically examine the correlates ofwhat they call succession outcomes, defined in terms of leader longevityand party results at the first post-succession general elections.

Paul Strangio’s chapter looks at life after succession. Studying the polit-ical afterlives of former Australian prime ministers, he examines the extentto which they suffer from ‘office dependency’ (since the loss of high officehardly means the individuals involved necessarily lose the desire to lead),continue to exert influence in their parties (and thus aid or nag their suc-cessors), and occupy significant public offices.

Paul Atkins’ chapter switches the focus from political to managerial suc-cessions. He laments the fact many successions within the senior ranks ofthe public service are marred by lost opportunities to build trust betweennew leaders and their staffs and stakeholders. To remedy this he offers amodel of ‘collaborative enquiry’ to facilitate trust-building and organiza-tional learning when a new leader comes into the job.

Bibliography

Alderman, R. K. (1999) ‘Revision of leadership election procedures in the conservativeparty’, Parliamentary Affairs, 52(2), 260–74.

Alderman, R. K. and Smith, M. J. (1990) ‘Can British prime ministers be given thepush by their parties?’, Parliamentary Affairs, 43(3), 260–72.

Alexandre, P. (1970) Le Duel: De Gaulle – Pompidou (Paris: Grasset).Alter, J. (2007) The Defining Moment: FDR’s Hundred Days and the Triumph of Hope

(New York: Simon & Schuster).Andrews, J. T. and Jackman, R. W. (2008) ‘If winning isn’t everything, why do they

keep score? Consequences of electoral performance for party leaders’, British Journal ofPolitical Science, 38(4), 657–75.

Baring, A. (1982) Machtwechsel: Die Ära Brandt-Scheel (Stuttgart: Deutsche Verlags-Anstallt).

Berlinski, S., Dewan, T. and Dowding, K. (2007) ‘The length of ministerial tenure inthe United Kingdom, 1945–1997’, British Journal of Political Science, 37(2), 245–62.


Berlinski, S., Dewan, T. and Dowding, K. (2010) ‘The impact of individual and collec-tive performance on ministerial tenure’, The Journal of Politics, 72(2), 559–71.

Bille, L. (1997) ‘Leadership change and party change: The case of the Danish SocialDemocratic Party, 1960–1995’, Party Politics, 3(3), 379–90.

Blondel, J. (1982) The Organization of Governments: A Comparative Analysis of Govern-mental Structures (London: Sage).

Borchert, J. and Zeiss, J. (eds) (2003) The Political Class in Advanced Democracies(Oxford: Oxford University Press).

Bueno de Mesquita, B., Smith, A., Siverson, R. M. and Morrow, J. D. (2003) The Logicof Political Survival (Cambridge, Mass.: MIT Press).

Bunce, V. (1981) Do New Leaders Make a Difference? (Princeton: Princeton UniversityPress).

Bynander, F. and ’t Hart, P. (2006) ‘When power changes hands: The political psycho-logy of leadership succession in democracies’, Political Psychology, 27(5), 707–29.

Bynander, F. and ’t Hart, P. (2007) ‘The politics of party leader survival and succession:Australia in comparative perspective’, Australian Journal of Political Science, 42(1),47–72.

Bynander, F. and ’t Hart, P. (2008) ‘The art of handing over: (Mis)managing partyleadership successions’, Government & Opposition, 43(2), 385–404.

Calvert, P. (ed.) (1987a) The Process of Political Succession (Houndmills: Macmillan).Calvert, P. (1987b) ‘Political succession in Western Europe’, in P. Calvert (ed.) The

Process of Political Succession (Houndmills: Macmillan), pp. 27–58.Colome, G. and Lopez Nieto, L. (1993) ‘The selection of party leaders in Spain:

Socialist cohesion and opposition turmoil’, European Journal of Political Research,24(3), 349–60.

Courtney, J. C. (1995) Do Conventions Matter? Choosing National Party Leaders inCanada (Montreal: McGill-Queens University Press).

Cross, W. and Blais, A. (2009) Who selects the party leader? A cross-national analysis.Unpublished Paper, Carleton University.

Davis, J. W. (1998) Leadership Selection in Six Western Democracies (Westport:Greenwood).

De Winter, L. (1993) ‘The selection of party presidents in Belgium: Rubber-stampingthe nominee of the party elites’, European Journal of Political Research, 24(3),233–56.

Denham, A. (2009) ‘Far from home: Conservative leadership selection from Heath toCameron’, The Political Quarterly, 80(3), 380–7.

Denham, A. and O’Hara, K. (2008) Democratising Conservative Leadership Selection:From Grey Suits to Grass Roots (Manchester: Manchester University Press).

Dewan, T. and Dowding, K. (2005) ‘The corrective effect of ministerial resignationson government popularity’, American Journal of Political Science, 49(1), 46–56.

Dogan, M. (ed.) (2003) Elite Configurations at the Apex of Power (Boston, Mass.: Brill).Dowding, K. and Dumont, P. (eds) (2008) The Selection of Ministers in Europe: Hiring

and Firing (London: Routledge).Dowding, K. and Kang, W.-T. (1998) ‘Ministerial resignations 1945–97’, Public

Administration, 76(3), 411–29.Elgie, R. (ed.) (1999) Semi-presidentialism in Europe (Oxford: Oxford University Press).George, A. L. and Bennett, A. (2005) Case Studies and Theory Development in the Social

Sciences (Cambridge, Mass.: MIT Press).Giambatista, R. C., Rowe, W. G. and Riaz, S. (2005) ‘Nothing succeeds like succession:

A critical review of leader succession literature since 1994’, The Leadership Quarterly,16(6), 963–91.


Goodin, R. E. and Tilly, C. (eds) (2006) The Oxford Handbook of Contextual PoliticalAnalysis (Oxford: Oxford University Press).

’t Hart, P. (2007) ‘How Adenauer lost power: Understanding leadership rivalry andsuccession conflict in political parties’, German Politics, 16(2), 273–91.

Heppell, T. (2008) Choosing the Tory Leader: Conservative Party Leadership Elections fromHeath to Cameron (London: Taurus).

Heppell, T. (2010) Choosing the Labour Leader: Labour Party Leadership Elections fromWilson to Brown (London: Taurus).

Hermann, M. (2005) ‘Who becomes a political leader? Leadership succession, gener-ational change, and foreign policy’, Paper, International Studies Association con-ference, <http://www.allacademic.com/meta/p69844_index.html>

Hood, C. C. and Peters, B. G. (eds) (1994) Rewards at the Top (London: Sage).Hood, C. C. and Lodge, M. (2006) The Politics of Public Service Bargains: Reward,

Competency, Loyalty and Blame (Oxford: Oxford University Press).Huson, M. R., Malatesta, P. H. and Parrino, R. (2004) ‘Managerial succession and firm

performance’, Journal of Financial Economics, 74(2), 237–75.Jackson, K. (1975) ‘Political leadership and succession in the New Zealand National

Party’, Political Science, 27(1/2), 1–24.Kane, J., Patapan, H. and ’t Hart, P. (eds) (2009) Dispersed Democratic Leadership:

Origins, Dynamics, and Implications (Oxford: Oxford University Press).Kelly, P. (1984) The Hawke Ascendancy: A Definitive Account of Its Origins and Climax

1975–1983 (London: Angus & Robertson).Kenig, O. (2008) ‘Democratization of party leadership selection: Do wider selec-

torates produce more competitive contests?’, Electoral Studies, 28(2), 240–7.Kenig, O. (2009) ‘Democratizing party leadership selection in Israel: A balance sheet’,

Israel Studies Forum, 24(1), 62–81.Kesner, I. F. and Sebora, T. C. (1994) ‘Executive succession: Past, present and future’,

Journal of Management, 20(2), 327–72.Koerfer, D. (1987) Kampf ums Kanzleramt: Erhard und Adenauer (Stuttgart: Deutsche

Verlags-Anstallt).LeDuc, L. (2001) ‘Democratizing party leadership selection’, Party Politics, 7(3), 323–41.Linz, J. J. and Stepan, A. (eds) (1978) The Breakdown of Democratic Regimes (Baltimore:

John Hopkins).Linz, J. J. and Stepan, A. (1996) Problems of Democratic Transition and Consolidation:

Southern Europe, South America and Post-Communist Europe (Baltimore: John Hopkins).Marsh, M. (1993) ‘Selecting party leaders in the Republic of Ireland: Taking the lid

off party politics’, European Journal of Political Research, 24(3), 295–316.McCauley, M. and Carter, S. (eds) (1986) Leadership and Succession in the Soviet Union,

Eastern Europe and China (London: Macmillan).Miliband, R. (1973) The State in Capitalist Society: The Analysis of the Western System of

Power (London: Quartet Books).Mills, C. W. (1956) The Power Elite (Oxford: Oxford University Press). Mulgan, R. (2003) Holding Power to Account (Basingstoke: Palgrave).Mulgan, R. (2007) ‘Truth in government and the politicization of public service

advice’, Public Administration, 85(3), 569–86.Mulgan, R. (2008) ‘How much responsiveness is too much or too little?’, Australian

Journal of Public Administration, 67(3), 345–56.Naughtie, J. (2002) The Rivals: Blair and Brown – The Intimate Story of a Political

Marriage (London: Fourth Estate).Ndofor, H. A., Priem, R. L., Rathburn, J. A. and Dhir, A. K. (2009) ‘What does the new

boss think?: How new leaders’ cognitive communities and recent “top-job” success


affect organizational change and performance’, The Leadership Quarterly, 20(5),799–813.

Offe, C. (1996) The Varieties of Transition (Cambridge, Mass.: MIT Press).Ornstein, N. J. and Mann, T. E. (eds) (2000) The Permanent Campaign and Its Future

(Washington, D.C.: American Enterprise Institute).Pennings, P. and Hazan, R. Y. (2001) ‘Democratizing candidate selection: Causes and

consequences’, Party Politics, 7(3), 267–75.Perry, J. L. and Hondeghem, A. (eds) (2008) Motivation in Public Management: The Call

of Public Service (Oxford: Oxford University Press).Podger, A. (2007a) ‘What really happens: Departmental secretary appointments, con-

tracts and performance pay in the Australian Public Service’, Australian Journal ofPublic Administration, 66(2), 131–47.

Podger, A. (2007b) ‘Response to Peter Shergold’, Australian Journal of Public Adminis-tration, 66(4), 498–500.

Punnett, R. (1992) Selecting the Party Leader: Britain in Comparative Perspective (London:Wheatsheaf).

Quinn, T. (2004) ‘Electing the leader: The British Labour Party’s electoral college’,British Journal of Politics and International Relations, 6(3), 333–52.

Quinn, T. (2005) ‘Leasehold or freehold? Leader-eviction rules in the British Conserv-ative and Labour parties’, Political Studies, 53(4), 793–815.

Raffel, J. A., Leisink, P. and Middlebrooks, A. E. (eds) (2009) Public Sector Leadership:International Challenges and Perspectives (Cheltenham, Glos.: Edward Elgar).

Rahat, G. and Hazan, R. Y. (2001) ‘Candidate selection methods: An analytical frame-work’, Party Politics, 7(3), 297–322.

Reid, A. (1969) The Power Struggle (Sydney: Shakespeare Head Press).Rhodes, R. (1997) Understanding Governance: Policy Networks, Governance, Reflexivity

and Accountability (Buckingham: Open University Press).Rhodes, R., Wanna, J. and Weller, P. (2009) Comparing Westminster (Oxford: Oxford

University Press).Rowe, W. G., Cannella Jr., A. A., Rankin, D. and Gorman, D. (2005) ‘Leader succession

and organizational performance: Integrating the common-sense, ritual scapegoating,and vicious-circle succession theories’, The Leadership Quarterly, 16(2), 197–219.

Savoie, D. J. (2003) Breaking the Bargain – Public Servants, Ministers and Parliament(Toronto: University of Toronto Press).

Shergold, P. (2007) ‘What really happens in the Australian Public Service: An alternativeview’, Australian Journal of Public Administration, 66(3), 367–70.

Stark, L. (1996) Choosing a Leader: Party Leadership Contests in Britain from MacMillanto Blair (London: Macmillan).

Strøm, K. (1993) ‘Competition ruins the good life: Party leadership in Norway’, EuropeanJournal of Political Research, 24(3), 317–47.

Swaine, M. (1992) The Military and Political Succession in China: Leadership, Institutions,Beliefs (Santa Monica: RAND).

Uhr, J. (2005) Terms of Trust (Sydney: UNSW Press).Weller, P. (1983) ‘The vulnerability of prime ministers: A comparative perspective’,

Parliamentary Affairs, 36(1), 96–117.Weller, P. (1985) First Among Equals (Sydney: Allen & Unwin).Weller, P. (1994) ‘Party rules and the dismissal of prime ministers: Comparative per-

spectives from Britain, Canada and Australia’, Parliamentary Affairs, 47(1), 133–43.Wong, J. and Yongnian, Z. (eds) (2002) China’s Post-Jiang Leadership Succession: Problems

and Perspectives (Singapore: Singapore University Press).


Part I

Understanding Transitions

2Ruling Elite Transitions in Australiaand the United StatesJohn Higley and Jan Pakulski

Introduction

In November 2007 Australian voters elected a new political leader, KevinRudd. His election was the culmination of a campaign focused almostentirely on the contrasting personalities and images of Rudd and JohnHoward. By projecting the image of a flexible technocrat-reformer repre-senting a new generation of Australians, Rudd triumphed over Howard’simage as a stubborn defender of the established political-economic orderand the generation it served. In November 2008 American voters alsoelected a new leader, Barack Obama, following a lengthy presidential cam-paign focused heavily on the personalities and images of Obama and hisopponent John McCain. Although affected in its final weeks by a shatter-ing financial crisis, the American contest posed a choice between a highlyeducated black man and a white military hero, between inspiration andtoughness.

We have elsewhere discussed Max Weber’s concept of ‘Leader Democracy’and its salience to Australia, the United States and other western demo-cracies (Pakulski and Higley, 2008). We regard the leader-centred elec-tion campaigns and victories of Rudd and Obama as illustrating thissalience once again. However, we believe it is also useful to view these and similar political transitions in wider perspective. A device for doing so is Vilfredo Pareto’s theory of what he calls cyclical successions by funda-mentally different types of ruling elites.1 He, and before him Machiavellimetaphorically labelled them ‘lions’ and ‘foxes’. Whereas Weber con-centrated on charismatic leaders and their plebiscitary relations with mass publics, Pareto focused on the ascendance and recession of rulingelites tout court. An understanding of political transitions benefits fromboth approaches, but here we interpret recent changes in the compositionof ruling elites within Australia and the US through the prism of Pareto’stheory.

23

Pareto’s theory of elite transitions

In his million-word Treatise on General Sociology, published in English as TheMind and Society (1916/1935), Pareto surveyed widely through western history,discussing many different societies, political regimes, and alternating eras offaith and scepticism. He did this in order to illustrate an elaborate and intri-cate theory of humanity’s several and conflicting psychological propensities(‘residues’), the rationalizations, superstitions, and ideologies that are derivedfrom them (‘derivations’), and how changing combinations of psychologicalpropensities and derived beliefs, together with concrete economic interests,shape the fortunes of ruling elites and societies. While there is loose familiaritywith Pareto’s theory, it has seldom been applied to actual events and pro-cesses. Compared with the attention showered on the theories of Marx andWeber or Schumpeter’s competitive theory of democracy (‘democratic elitism’),Pareto’s theory has received short shrift (see Finer, 1966, 1968; Femia, 2006;Marshall, 2007; Best and Higley, 2009).

Pareto conceived of ruling elites as broad and complex alliances of diversepolitical groups, economic interests, and social forces, the inner leadershipsof which are located in parliaments and governments. In their modernform, ruling elites contain parties and other factions that compete for politicaland economic power, squabble endlessly over policy matters, and rotate inand out of government. But Pareto viewed such competitions, squabbles,and rotations as matters that do not alter the basic psychology and struc-ture of a ruling elite. Psychologically, its members are disposed to combinethe two modes of political rule, force and persuasion (or coercion and cun-ning), although over time a disposition to rely on just one of the twomodes predominates (Femia, 2006: 70). Structurally, elite members consti-tute an intertwined and polyarchal web of patrons with diverse clienteles.A ruling elite is never homogeneous or conspiratorial; it is a far-flung orderor system in which a ‘common accord’ results from ‘an infinitude of minoracts, each determined by … present advantage’ (Pareto, 1916/1935: 2254).In S. E. Finer’s apt formulation, the members of a ruling elite ‘live by takingin each others’ washing’, with the elite’s inner leadership ‘looking after’ asmany patrons and clienteles as possible (1968: 447; see also Femia, 2006:107–8). Thus Pareto regarded opposing parties such as America’s Repub-licans and Democrats, or Australia’s Liberals and Labor, together with theirrespective business, trade union and numerous other allies, as of a piece – as constituting a ruling elite (Pareto, 1916/1935: 2257–8). But, he added,there are always other individuals and groups hostile to the existing elite’smode of rule waiting to take power in some crisis situation. In this respect,Pareto regarded the durations of ruling elites as coterminous with political-economic eras.

As his fox and lion metaphors suggest, Pareto postulated that ruling elitesdiffer according to which of two underlying psychological propensities pre-

24 Ruling Elite Transitions in Australia and the United States

dominates over time: (1) a propensity for combining things innovativelythat is characteristic of fox-like politicians and their entrepreneurial (‘spec-ulator’) economic allies; (2) a propensity for restoring things to traditionalforms that is characteristic of lion-like politicians and their fixed-income(‘rentier’) economic allies. Members of a fox-speculator alliance tend to be calculating, deceptive, and inventive; those making up a lion-rentieralliance tend to be belligerent, courageous, and determined (Femia, 2006:105). Readers of Pareto have usually regarded these postulates as more orless metaphysical, and his theory of cyclically alternating elites has con-sequently been seen as not more than an inspired guess lacking empiricalsupport (Finer, 1966: 84). However, research in modern political psychology,ably synthesized by Alasdair Marshall (2007), lends Pareto’s distinctionsempirical credibility. It uncovers a cluster of personality traits consisting ofhedonistic, irreligious, tolerant, and risk-taking proclivities that resemblethe propensity Pareto assigned to fox-speculator elites, and an opposite clus-ter of risk aversion, paranoia, anti-introspection, and obsessional prejudiceapproximating the propensity he assigned to lion-rentier elites.

The centrepiece of Pareto’s theory is his contention that ruling elite circulation, degeneration, and transition are sequential stages in any political-economic era. A ruling elite is never static; its composition undergoes con-stant, albeit usually slow and incremental change, with new members risingfrom non-elite statuses and replacing old elite members gradually. This revi-talizes the elite by bringing in fresh blood and talent. Although Pareto did notclearly say so, elections are a key modern mechanism of circulation within aruling elite. In the American and Australian contexts, Democrats or Repub-licans, Liberals or Labor win elections and bring new persons and allies intoelite positions, though they do so without changing existing policy mixes and sets of belief in any fundamental way. In certain era-ending crises, how-ever, this routine elite circulation gives way to more abrupt and sweeping elitetransitions that mark revolutions or gross economic and political calamities(Pareto, 1916/1935: 2054–6). Such transitions are the consequence of inexor-able elite degeneration and the increasingly dire situation it creates.

Ruling elite degeneration occurs in two interrelated ways according toPareto. First, the elite’s intellectual and moral qualities decline, making itsmembers and especially its leaders increasingly prone to biases, miscalcula-tions, and policy blunders. Second, its rule becomes too one-sided, renderingleaders and members less and less capable of responding effectively to chan-ging circumstances and challenges (1916/1935: 2178). The former processresults principally from elite closure and faulty recruitment. Able persons whodo not fit the elite’s social profile and intellectual fixations are increasinglykept out of its ranks, with recruitment more and more dominated by inherit-ance, cronyism and sycophancy. The other process – increasing one-sidednessand rigidity – involves a growing predominance, especially in the leadershipcore, of one personality type, whether leonine or vulpine. This fosters a

John Higley and Jan Pakulski 25

biased repertoire of policies and strategies in place of balanced responses to changing circumstances and challenges – it inhibits being forceful andpersuasive, principled and opportunistic.

More specifically, an elite in which leonine propensities come to pre-dominate fastens ever more strongly on conservative, nationalistic, andreligious recipes and shibboleths that bias its policies and actions in a sys-tematic way. It invokes some natural order of things, such as an economy’s‘invisible hand’ or ‘God’s will,’ to justify inadequately funded domestic policies and international adventures that consume accumulated wealth,deplete military and other resources, and undermine political stability.Growing indebtedness, combined with favouring entrenched over entre-preneurial business interests, weakens the economy and sooner or later createsa crisis. Conversely, an elite that has predominantly vulpine propensities reliesincreasingly on reason rather than faith, value diversity and innovation ratherthan stable norms and traditions. It pursues a crafty politics involving end-less deals and payoffs that gradually dissipate the elite’s authority and fostersocial decadence. The elite’s incessant efforts to placate its many clientelesfeed economic inflation and eventual insolvency (Femia, 2006: 71–2). Wherechanging circumstances require blunt, courageous, and forceful actions, a vulpine elite’s systematic bias toward deals and payoffs becomes a fatalliability.

Thus both types of ruling elites degenerate, especially in their core leader-ships, and this culminates in crises that herald elite transitions. These maytake violent forms, as in revolutions or military coups, or they may involvethe thorough discrediting of an elite’s ability to cope with crisis. In thelatter situation, individuals and groups hostile to the elite’s predominantmode of rule out-manoeuvre it politically. The ruling elite is repudiated insome relatively comprehensive way and many of its members lose power.

Pareto died in 1923 and it was principally during the last few years of hislife that he focused on ‘democracies’ (1921/1984). He contended that theyinvariably contain ruling elites: ‘Whether universal suffrage prevails or not,it is always an oligarchy that governs’ (1916/1935: 2183). A democracy is in reality a ‘demagogic plutocracy’ ruled through bribes, deceptions anddemagogy by fox-like politicians and entrepreneurial speculators. But becausethis rule gradually allocates more wealth than it creates, it eventually ‘killsthe goose that lays the golden eggs’ (Finer, 1966: 142). When the goose isdead – when the demagogic plutocracy is effectively insolvent – a crisisoccurs and in it more lion-like politicians and rentiers living off inherit-ances and investments gain the upper hand. They rely increasingly on coer-cion to reverse the economic insolvency and social decadence to whichfox-speculator rule led. A ‘military plutocracy’ may be created. But the lion-rentier ruling elite also gradually degenerates and the political-economicorder again heads toward crisis, not least because of the elite’s unsustain-able ‘warlike activities’. In this next crisis, fox-speculators again take over,


re-institute a demagogic plutocracy, and initiate a new era – a new ‘pluto-cratic cycle’ (1921/1984: 55–62; see Finer, 1968; Powers, 1984: 1–25; Femia,2006: 100–23).

It is obvious that Pareto painted with a broad brush. He was fully awarethat the contours and actions of ruling elites differ greatly between societiesand eras and that in its details history never repeats itself. He contended,nonetheless, that uniformities and tendencies, regularities and patterns arediscernible among historically specific cases. He knew that the underlyingpropensities he assigned to ruling elites are theoretical constructs thatcannot be validated through direct observation and measurement. But hisgoal was a general theory that could account better than competing theo-ries – Pareto regarded Marxism as the principal competitor – for sweepingand undulating political-economic changes in history. He believed thatthese changes could best be explained in terms of the elite propensities anddegenerative processes he postulated. Joseph Femia (2006: 75) neatly sum-marizes Pareto’s project and method:

[His] message is clear … In order to discover the ‘structure’ (or ‘edifice’)of the historical process, we must resort to abstraction and accept thatdifferent phenomena will be encompassed by a single category, thoughit is actually a composite of numerous elements – in the same way that ageologist gives the name ‘clay’ to a compound of a number of chemicalelements.

Let us ask if Pareto’s theory illuminates recent political successions in Australiaand the United States.

A leonine elite ascendancy, 1980–2008

A mountain of books and articles recount the comings and goings, suc-cesses and failures, of political leaders, governments, and other prominentpersonalities and elite groups in Australia and the US during the past 30 years. But studies of how the two countries’ ruling elites, in Pareto’sencompassing sense, have waxed and waned are to our knowledge non-existent (although G. William Domhoff’s repeatedly updated and empiricallyrich studies of a US ‘ruling class’ are a good approximation; see Domhoff,2009). Here we interpret features and trends in ways that accord with Pareto’stheory. We observe that the past three decades entailed rule by increasinglyaggressive and tough-minded elites resembling Pareto’s leonine type, thoughthis was more transparently so in the US than Australia. A leonine proclivitywas discernible from the period’s start, especially in the US, but for a whileelites in both countries balanced force and persuasion with considerable effect-iveness. However, from the mid-1990s the leonine proclivity became moreevident, with elite actions and assessments guided ever more rigidly by faith


in neo-liberal shibboleths at home and the efficacy of military force abroad.Viewed through Pareto’s prism, degenerations of ruling American and Austra-lian elites accelerated and led eventually to crises and elite transitions.

Leonine ascendancy in the United States

Profiting politically from post-Vietnam economic and political disarray at the end of the 1970s, an elite faction headed by Ronald Reagan gainedexecutive power in Washington at the start of 1981 and was widely viewedas instituting a basic change in governing style and policy direction – whatcame to be termed ‘Reaganism’ and ‘Reaganomics’ (Wilentz, 2008). A com-parable change in governing style and policy direction at the start of the1980s was less evident in Australia, even though the country had suffered adeclining economic trajectory during the 1970s and a constitutional crisisthat punctuated the decline in 1975. The Liberal faction that dominatedgovernment from 1975 until 1983 did not depart in a marked way, other thansome tougher rhetoric, from what had gone before. Led by Bob Hawke, theLabor faction that gained power in 1983 recognized that the post-World WarII economic order was no longer viable and did much to alter it (Kelly, 1992).Still, in both the US and Australia changes in elites from the late 1970s to theearly 1980s could be regarded as customary short-term political shifts acrossthe centre-right/centre-left divide that resulted mainly from the charismaticqualities of Reagan and Hawke, their predecessors’ vacillations, and theiropponents’ colourlessness.

Perhaps, however, these were surface appearances. Clustered around Reaganwere belligerent nationalist and neo-conservative elite factions determined to‘unshackle’ the economy and restore America’s military power in the wake ofthe Vietnam debacle. During the Reagan administration’s eight years, spend-ing on the military and advanced weapons systems soared; the intelligenceapparatus was expanded to gigantic size; the Soviet Union was labelled an‘Evil Empire’ and a ‘Star Wars’ challenge to it was mounted; tiny Grenada wasinvaded to unseat a Marxist government; air strikes on Libya were launched inretaliation for a terrorist attack in Germany; arms were supplied to mujaheddininsurgents in Afghanistan and rebel ‘contras’ in Nicaragua; staunch economicand political support of Israel and other allies, many of them autocratic, wasrendered. Proclaiming in domestic matters that ‘Government is the problem,not the solution’, the Reagan administration cut taxes on the well-off, con-fronted trade unions, and supported the Federal Reserve’s deliberate instiga-tion of a deep recession to wring inflation out of the economy. In 1982 theadministration secured congressional passage of the Garn-St. GermainDepository Institutions Act, which greatly relaxed regulation of Savings andLoans institutions, enabling them to gamble with taxpayer money becausethe federal government guaranteed ‘S&L’ deposits. This led to hundreds of‘S&L’ insolvencies that cost $130 billion in public funds – until then the cost-liest financial scandal in American history (Patterson, 2005: 175). Important


for the longer run, the Reagan administration lessened restrictions on mort-gage lending, thus enabling home buyers to obtain mortgages without havingto make large equity down payments. This planted the seeds of an eventualhousing bubble and a precipitous decline in consumer savings.

During the succeeding presidency of George Bush senior, which was wonin a ruthless campaign against his Democratic opponent Michael Dukakis,leaders of the nationalist and neo-conservative factions held high offices innational security and economic policy domains, the most prominent beingDick Cheney and Paul Wolfowitz as Secretary and Deputy Secretary of Defence,respectively. However, the Bush administration offset part of the huge Reaganbudget deficits and economic risk-taking by increasing government tax rev-enues. Apart from a brief intervention in Panama to remove a dictator andprotect US control of the Panama Canal, it eschewed unilateral military actionsabroad, as illustrated by the large multinational force it assembled to driveIraqi forces out of Kuwait early in 1991.

Courtesy of a political interloper, H. Ross Perot, who captured 19 per cent ofthe popular vote in the 1992 presidential election, Bill Clinton and the Demo-crats edged into the White House. Their hold on Congress was tenuous fromthe start, and it was lost in 1994 mid-term elections when Republicans, headedby a gaggle of aggressively pious, nationalist, and obsessively neo-liberal south-ern newcomers, gained what proved to be lasting control of Congress. Refusingto compromise on the government cutbacks they had promised, the con-gressional Republicans forced a brief and partial federal government shut-down in late 1995. Three years later House Republicans impeached Clinton,though their Senate colleagues lacked the two-thirds majority needed to con-vict him. In 1999 the Republican-controlled Congress passed the Gramm-Leach-Biley Act. This repealed that part of the Glass-Steagall Act, passed duringthe 1930s, which prohibited mergers and direct competitions between banks,insurance, investment, and security-trading firms. The new Act also exemptedsecurities-based swap agreements from Security and Exchange Commissionoversight. In these and other actions, the Republican project of unshacklingmarkets looked very much like de-regulation for its own sake. It produced‘exuberant’ – many would now say wild – financial risk-taking and hugeprofits for the Republicans’ securities-trading allies on Wall Street.

A slide toward crisis accelerated following the victory of an alliance ofdogmatically neo-liberal and neo-conservative elite factions, headed by theunworldly George W. Bush, in the 2000 presidential election. Bush’s victory,like his father’s in 1988, was gained through ruthless actions during an elec-toral standoff in Florida. Once in command of the White House, the Bushcoterie systematically filled government posts with loyal and like-mindedassociates, a practice that culminated after Bush’s re-election in 2004 whentrusted friends and close associates were appointed to nine of the 16 cabinetpositions and Bush nominated a Texan crony, Harriet Meiers, for the SupremeCourt. The 9/11 attacks on New York and Washington DC in 2001 enabled


the administration to break free of most political constraints, and within amonth it regrouped around an inner circle calling itself ‘the Vulcans’ (Mann,2004). A retaliatory military attack on Taliban-controlled Afghanistan waslaunched immediately, and not later than July 2002 Bush and his closest asso-ciates decided to invade Iraq and remove Saddam Hussein from power, a unilateral undertaking that Bush regarded as approved by a ‘higher father’than his own (Woodward, 2004: 421).

Contrary to traditional Republican beliefs in limited government, balancedbudgets, and tight monetary policy, but in line with Bush’s intuitions and thebeliefs of neo-liberals and neo-conservatives around him (Kaplan, 2008), mil-itary spending was again greatly boosted and the Iraq adventure came on theheels of massive tax reductions for the well-off. Budget deficits much greaterthan those of the Reagan administration resulted, and they had to be coveredby borrowing funds from abroad, eventually at the rate of $1 billion per day.In monetary policy, Alan Greenspan, the Federal Reserve chairman and archfree-marketeer, lowered the basic federal interest rate close to zero and kept itthere. Budget deficits and the low interest rate, combined with a strict hands-off posture toward financial markets, especially the market for sub-prime (i.e.high-risk) mortgages, powered a spectacular stock market rise and housingbubble, along with a consumer spending spree that eventually left the averagehousehold owing $10,000 in credit card debt. The cocky view of the admin-istration and its business allies about these developments was encapsulated by Vice President Cheney’s flip remark: ‘Reagan proved deficits don’t matter’(Phillips, 2006: 323). Together with a vicious attack on the military record ofBush’s 2004 presidential opponent, John Kerry, the deficit-stoked economyand consequent good times enabled Bush to win re-election handily.

But by autumn 2008, following an ominous bankruptcy and $30 billiongovernment bailout of the Bear Stearns investment firm earlier in the year(Cohan, 2009), the Bush administration was facing economic Armageddon.The federal budget deficit for FY 08-09 was spiralling beyond $1 trillion;two huge government-backed mortgage lending agencies, Fannie Mae andFreddie Mac, had to be de facto nationalized; the major investment firmLehman Brothers went to the wall with shattering ramifications; the AIGinsurance company, which had worldwide financial ties and obligations, hadto be bailed out with some $182 billion in taxpayer funds; General Motorsrequired a government infusion of $50 billion to avoid immediate bankruptcyand the direct and indirect loss of perhaps 200,000 jobs; Chrysler was savedonly by the government forcing it to amalgamate with Fiat; and staving off the collapse of nearly all major banks required ‘temporary relief funds’ in exchange for government equity holdings. The overall cost was at least$800 billion in government funds, with an additional $1 trillion injected intothe economy by the Federal Reserve. Stock market, investment, and otherlosses totalled $4 trillion. These and other fevered efforts to prevent the onsetof economic depression (see Wessel, 2009) occurred at the height of the 2008


presidential election campaign. The crisis demolished the credibility of John McCain and ensured Barack Obama’s victory, an outcome that probablywould not have occurred had it not been for the crisis (Linn et al., 2009).

Leonine ascendancy in Australia

In Australia a more leonine style of elite rule was presaged by the rise of factions associated initially with two political leaders, John Hewson in the Liberal Party and Paul Keating in the Labor Party. Hewson proclaimedunbounded faith in ‘economic rationalism’, a label for the doctrinaire laissez-faire capitalism that pervaded a considerable part of his party and its support-ers in business and intellectual circles as an antidote to the electoral successesof Bob Hawke and Labor in presiding over a mixed economy during the1980s. However, Hewson fell on his electoral sword in 1993 by promising to introduce a regressive Goods and Services Tax and vowing to reduce theCommonwealth government to a bare minimum of regulatory and wel-fare functions. Keating, a pugnacious product of the hard-edged Labor Partymachine in New South Wales, had done much as Hawke’s Treasurer to de-regulate parts of the economy. In that capacity he initiated a decade of toughLabor reforms, reducing the budget deficit (AU$10bn in 1983), introducingenterprise wage bargaining, floating the Australian dollar, lowering taxes andimport tariffs, and – against strong opposition within his party – privatizinglarge state corporations, including Qantas and Commonwealth Bank. In late1991 Keating engineered Hawke’s overthrow and as prime minister during the next four years he articulated a muscular Australian nationalism thatreviled the country’s servitude to Britain and the US, demanded that Australiabecome a republic, and depicted its destiny as lying squarely in Asia. He alsooversaw the mandatory detention of asylum seekers, championed further taxreform, and took a tough stand on the Timor Gap Treaty, thus acquiring theimage of a ‘dry’ reformer who paved the way for John Howard (Love, 2008).Keating and the loyalists around him bore a stylistic resemblance to thesouthern-led Republicans who held sway in the US after 1994.

Domination of Australia’s politics by Keating and his faction, though con-sequential, was brief. In 1996, John Howard, who had regained leadership of the Liberal Party, defeated Keating and began an exceptionally long tenureas prime minister. Protected economically by rapidly expanding markets forAustralia’s commodity exports and politically by uninspiring or erratic oppo-sition leaders, Howard and his faction ruled in a markedly aggressive wayduring four election cycles. They committed military forces to the US-ledinvasions of Afghanistan and Iraq; intervened militarily in East Timor and theSolomon Islands; instituted invasive anti-terrorist laws and surveillance prac-tices; used a ‘Pacific Solution’ and harsh detention camps to deter asylumseekers; privatized and out-sourced many government functions; alteredindustrial relations to give employers more leeway; cracked down on welfaresupport for the indigent and unemployed. Few Australians doubted the


Howard elite’s tough-minded character and firm beliefs in defence of family,national security, law and order, social unity, individual responsibility, and alightly regulated capitalist economy (Errington and van Onselen, 2007: 398).

The Howard elite was stable, close-knit, and well linked with strategic econ-omic interests. At its core was a handful of experienced and loyal Howardfriends and confidants. The Treasurer and Deputy Leader of the Liberal Party,Peter Costello, provided a strong connection with business lobbies and the H. R. Nicholls Society, an influential conservative think-tank. AlexanderDowner, scion of one of Australia’s most prominent political-aristocratic fam-ilies, linked the elite with old pastoral wealth and the social establishment; as minister for foreign affairs, Downer was fiercely pro-US. Philip Ruddock, the Attorney General and veteran MP, was an experienced and tough operatorin legal-administrative affairs and key architect of hard-line immigration pol-icies, especially border protection and mandatory detention of illegal migrants(a measure originally put in place by the Keating government). Tony Abbott,an ex-Rhodes scholar, amateur boxer and Howard protégé, had strong con-nections to the Catholic Church and the somewhat amorphous ‘moral major-ity’; he was an antagonist of trade unions and a vocal exponent of moralcauses such as anti-abortion. Another Howard favourite, Brendan Nelson, wasinitially given the task of de-regulating the education system, regarded byHoward and his associates as a haven for vulpine decadence.

Leonine proclivities became steadily more apparent. The Howard govern-ment won re-election in 1998 by positioning itself as the defender of middle-class ‘battlers’ and foe of multicultural and Aboriginal land claim excesses.Re-election in 2001 was secured by exploiting two events: the 9/11 attacks inthe US and the national security fears they aroused in Australia; and theTampa affair, in which the Howard government sent an SAS combat team to board a Norwegian vessel that had rescued several hundred Middle Eastrefugees headed for Australia in order to prevent them from being landed onAustralian territory and applying for legal refugee or asylum statuses. Duringthe 2004 election campaign a terrorist bomb wrecked the Australian embassyin Jakarta and assisted the Howard elite in mirroring the Bush elite’s success-ful 2004 re-election image as a tough defender of national security againstunreliable and weak-kneed opponents.

Compared with the US pattern, however, it is difficult to discern a cleardegeneration of Australia’s ruling elite after the mid-1990s. As noted, strongexport markets, together with a dazzling stock market rise, buoyed theHoward elite. By 2007, when sharply higher housing values levelled off andbegan to decline, there was a spreading sense that the elite had played outits hand. Still, nothing like the litany of failures that beset the Bush eliteafter 2004 occurred, and Australia’s increased economic turbulence resultedmainly from the deteriorating US economic situation. Although accom-panied by exceptional campaign fervour, the Howard elite’s defeat by KevinRudd and the resurgent Labor elite at the end of 2007 lay within the para-


meters of a normal political alternation, in which a majority of voters tire of a long-dominant faction and plump for a more youthful and energetic alter-native. If a comparatively indistinct leonine elite cycle ended in Australia itwas less because ‘lions’ overreached in fatal ways than because any significantaccomplishment achieved by Australia’s unpopular military involvement inIraq was impossible to demonstrate, while mounting US economic difficultiesdiscredited nostrums about linkages between limited government, unfetteredmarkets, and unending economic prosperity.

New vulpine ascendancies?

Pareto held that in ‘demagogic plutocracies’, such as Australia and theUnited States, politics are in largest measure battles for power between fac-tions within ruling elites (in terms of this volume: succession struggles).Individual leaders are crucial to the outcomes, and in key respects Australiaand the US have become leader democracies. Yet this may not be the wholestory. It is obvious that governing and opposing factions and their power-ful allies in business, trade unions, media, and assorted other organizedinterests constitute elite armadas, each of which possesses huge financialand other resources for waging political battles. Consider that the Obamaelite enlisted 300 officially designated foreign policy advisors, expendedthree quarters of a billion dollars to pay full-time organizers in most ofAmerica’s 3000 electoral districts and saturate commercial television andthe Internet with appeals for support, and appointed some 7000 persons tofederal government positions.

As the Obama efforts indicated, voters decide which leaders and affiliatedelites will hold government power, but voters’ decisions are tightly cir-cumscribed and manipulated. Pareto stressed that by taking advantage ofchanging circumstances one set of leaders and elites prevails by presentingitself as tougher, shrewder and more innovative than its opponents. But healso held that beneath the glitz of such electoral posturing competingleaders and elites are propelled – predominantly but never wholly – by thesame deep-seated psychological propensity and broad set of derived beliefs.Another way of putting this is to say that during an historical era politicalcontests consist much more of echoes than basic alternatives. In a timewhen the study of politics focuses on the behaviour of citizens and votersand tends to regard elites as blemishes on democracy, it is important torevive Pareto’s vision. But it is also important not to beg questions stem-ming from his vision. Does Pareto’s portrayal of ruling elites as lions andfoxes and of politics as continual cycles in their rise and fall make bettersense of political change than still more studies of voters, civil society,political trust, political culture and the like? In this chapter’s context, canwe, through Pareto’s prism, discern the start of another plutocratic cycle inwhich vulpine elites are again becoming ascendant?


In the United States veterans of the Clinton elite, whose hands were tiedby Republican control of Congress (and the Supreme Court) after 1994, filla considerable part of the Obama administration’s top echelons. Moreover,the 2008 elections brought substantial change in the makeup of Congress,and the economic crisis that opened the door to this change shredded neo-liberal orthodoxies. The Republican elite’s base of support has shrunk tosouthern and western states, most of them lightly populated. In keepingwith Pareto, the elite clustered around Obama displays some vulpine ear-marks. The craftiness with which Obama and his team bested nine othercandidates, including Hillary Clinton, for the Democratic nomination wasone. Another was Obama’s posture as a calm and measured alternative to the more hot headed McCain and the arrogant president, Bush. Where,for example, McCain stressed the necessity for open-ended US occupationof Iraq, Obama promised early withdrawal.

During its first half year in power the Obama elite displayed other vulpineearmarks. One was the dexterity with which it avoided making a direct andpolitically divisive assault on the preceding Bush administration, withObama and his lieutenants stating repeatedly that they saw no point inlooking back, they only wanted to look forward. Having promised to ‘endthe war in Iraq,’ Obama and his national security advisors took pains toassess the perils of US withdrawal and insist that military commanders’judgements would be decisive. Acceding to those judgements, the adminis-tration postponed major US withdrawal for at least a year. Meanwhile, itstepped up the military campaign in Afghanistan and missile strikes byunpiloted drones in northwest Pakistan, albeit in closer partnership withNATO and other forces and with greater emphasis on regional solutions.Nationalism was toned down, and the Bush elite’s invocations of America’smission in a ‘global war on terror’ were assiduously avoided. More broadly,diplomacy was back in favour with a focus on consensus building, nego-tiation, and working with ‘stake-holders’.

The Obama administration grappled with the economic crisis by eschewingradical and politically explosive measures such as bank nationalization andthe full business bankruptcies that neo-liberalism prescribes as the naturalsolution for insolvency. In short, the new governing elite acted in a prevail-ingly cautious and deliberate manner. To the extent that this presaged acontinuing pattern, an elite transition from aggressive and intransigent ruleto a more calculating and cunning mode occurred and perhaps signalledthe start of a new era.

Despite our doubt that similarities between US and Australian elite tra-jectories and transitions have been close, it is not hard to discern somefamily resemblances between the Obama and Rudd elites. The Rudd elite’selection campaign during 2007 had a crafty texture, playing up Howard’sbelligerence and hubris while offering a more comfortable yet competentand shrewd governing style. After winning office, Rudd and his associates


made eagerly awaited but largely cost-free gestures intended to signify a basicchange: an immediate ‘apology’ to indigenous Australians for past wrongs;signing the Kyoto Protocol to highlight awareness of climate change; Rudd’svisit to Beijing and his speech there in Mandarin. More costly policy changeswere, meanwhile, put on hold pending the completion of myriad reviews,though close to the end of its first election cycle, the Rudd government, likethe Obama administration before it, took on the major challenge of healthcare reform.

The question is whether the new US elite will survive long-lasting economicrecession. Measures taken to abate it will inevitably be deemed insufficientand widen elite divisions. Although neither set of governors appeared in mid-2009 to face a credible opposing elite with an attractive leader, the high andstill rising level of US unemployment was sucking wind rapidly from theObama sails, stepped-up combat in Afghanistan was becoming more costly insoldiers’ lives, and a political struggle over reform of the gigantic but deeplydysfunctional US health care system raged. The Rudd government did notface the military and political rigours of future withdrawal from Iraq and it was much less mired in the Afghan-Pakistan swamp; Chinese and otherdemand for Australian exports was reviving; and it appeared that an ambi-tious but politically difficult plan to cap and trade carbon emissions might beput, apart from window dressing, on the shelf to await international develop-ments. In the meantime a tougher test of its vulpine credentials was posed bythe resurgence within Australian politics of an issue that helped the Howardelite consolidate and strengthen its leonine rule: border control. Though theabsolute numbers were still small when compared internationally, a rapidgrowth in the number of refugees trying to reach the Australian continent illegally through people smugglers put the Rudd regime in a difficult spot. It had humanized the previously draconian border control regime, and nowfound itself faced with an increasingly nervous domestic public opinion and astrengthening Liberal Party opposition seeking to paint the government asweak and naïve in matters of national sovereignty and security.

All in all, at the time of writing more vulpine elites in both countries,though not as yet decisively marked, were being staunchly opposed by adher-ents to the preceding leonite ancien regimes. Still struggling to consolidate theirrule, they nevertheless looked like gradually getting into a position to be col-ouring political agendas and public policy during the next few years andperhaps beyond.

Note

1 Given that when using the term ‘elite successions’ Pareto actually refers to collec-tive turnover at the top government/society and not rotation of individual office-holders, we will use the term ‘transitions’ instead, in keeping with the conceptualdistinctions made in the introductory chapter in this volume.


Bibliography

Best, H. and Higley, J. (eds) (2009) ‘Democratic elitism reappraised’. Special issue ofComparative Sociology, 8(3), 323–44.

Cohan, W. D. (2009) House of Cards: A Tale of Hubris and Wretched Excess on Wall Street(New York: Doubleday).

Domhoff, G. W. (2009) Who Rules America? Challenges to Corporate and Class Dominance,6th edn (New York: McGraw Hill).

Errington, W. and van Onselen, P. (2007) John Winston Howard: The Biography(Melbourne: Melbourne University Press).

Femia, J. V. (2006) Pareto and Political Theory (London: Routledge).Finer, S. E. (1966) Vilfredo Pareto: Sociological Writings (New York: Praeger).Finer, S. E. (1968) ‘Pareto and pluto-democracy: The retreat to the Galapagos’, American

Political Science Review, 62(2), 440–50.Kaplan, F. (2008) Daydream Believers: How a Few Grand Ideas Wrecked American Power

(New York: Wiley).Kelly, P. (1992) The End of Certainty: The Story of the 1980’s (Sydney: Allen & Unwin).Linn, S., Moody, J. and Asper, S. (2009) ‘Explaining the horse race of 2008’, PS: Political

Science & Politics, 42(3), 459–66.Love, D. (2008) Unfinished Business: Paul Keating’s Interrupted Revolution (Melbourne:

Scribe).Mann, J. (2004) Rise of the Vulcans: The History of Bush’s War Cabinet (New York:

Viking).Marshall, A. J. (2007) Vilfredo Pareto’s Sociology: A Framework for Political Psychology

(Aldershot: Ashgate).Pakulski, J. and Higley, J. (2008) ‘Towards leader democracy?’, in P. ’t Hart and J. Uhr

(eds) Public Leadership: Perspectives and Practices (Canberra: ANU E Press), pp. 45–56.Pareto, V. (1916/1935) The Mind and Society, 4 vols (New York: Dover Publications).Pareto, V. (1921/1984) The Transformation of Democracy, edited by C. H. Powers

(New Brunswick: Transaction Publishers).Patterson, J. T. (2005) Restless Giant: The United States from Watergate to Bush v. Gore

(New York: Oxford University Press).Phillips, K. (2006) American Theocracy: The Peril and Politics of Radical Religion, Oil, and

Borrowed Money in the 21st Century (New York: Viking).Powers, C. H. (1984) ‘Introduction’, in V. Pareto, The Transformation of Democracy,

edited by C. H. Powers (New Brunswick, NJ: Transaction Publishers), pp. 1–23.Wessel, D. (2009) In Fed We Trust: Ben Bernanke’s War on the Great Panic (New York:

Harper).Wilentz, S. (2008) The Age of Reagan: A History 1974–2008 (New York: Harper). Woodward, R. (2004) Plan of Attack (New York: Simon & Schuster).


3Managers or Messiahs? PrimeMinisterial Leadership and theTransition to GovernmentJames Walter

Styles of prime ministerial transition management

The dichotomy between presidential and Westminster systems producesdistinctive interpretations of governmental transitions in Anglophone demo-cracies. When all significant executive positions fall to the winning party,as in the US, there must be more attention to the details of transition thanis required for a change of government in a parliamentary system, wherethe permanent civil service assures some continuity. Thus, the literature onpresidential transitions is more extensive, and more systematic, than thaton prime ministerial change (Burke, 2000; Kumar and Sullivan, 2003). Yet,over nearly 40 years, as a modernization of the Westminster executive hasbeen imposed by the necessities of rapid start-ups and decisive leadershipin a globalizing world, aspirant prime ministers have increasingly focusedon transition planning, in contrast to the ad hocery of the past. So therehave emerged some useful narratives on the preparation and transition topower of prime ministers, usually in the form of biography and memoir,and especially relating to those deemed to have driven change: see forinstance, studies of Thatcher and Blair on the path to power (Young, 1990,ch. 8; Hennessy, 1997; Hennessy, 2001: 476–86). Yet such studies are rarelycumulative or systematic – we have to read across them to detect a learn-ing curve, to map patterns or to attempt comparison with other cases of governmental transition.

This chapter surveys recent Australian prime ministers bridging the gapbetween opposition and government as an instance of systematic compar-ison. While focused specifically on the Australian case, it should provokereflection on parallels elsewhere and thus broader consideration of primeministerial transitions. It concentrates on the period from 1972 to 2007,since this has increasingly come to be seen as the era of contemporaryAustralian modernization (Kelly, 2009). It deals only with the transitions topower from opposition; hence, Paul Keating (Prime minister from late

37

1991–96), who assumed power while his party was in government, is excluded.It indicates not only that there is a linear development, as each incomingadministration adapts from what has gone before, but also that there are pat-terns distinctive to the dominant parties. Finally, it suggests that while cumu-lative patterns can be discerned, the intersections of a leader’s personal style,historical context and party fortunes reveal the influence of personal pre-dilections on these patterns. A spectrum of possible approaches, from zealouspursuit of a programme (salvation) at one end (Gough Whitlam, 1972–75),through appeals to values rather than policy (John Howard, 1996–2007),towards consensus building and change management (Bob Hawke, 1983–91)at the other end, is outlined. Whitlam’s reference to the importance of ‘theword’ in explaining his transition to government – he cites his party’s plat-form as the ‘Old Testament’ and his 1972 policy speech itself as the ‘NewTestament’ (Whitlam, 1985: 23) – prompts ‘messiah’ as designating one endof the spectrum. Hawke’s success in separating his own ‘love affair with theAustralian people’ from his managerial approach to a high achieving reformgovernment suggests ‘manager’ as the other. This chapter concludes that‘managers’ emerge as the most effective and productive transition leaders,even in situations that demand significant reform.

Preparation and transition: Labor’s experiences

The period of contemporary modernization was kicked off by the con-troversial Labor prime minister, Gough Whitlam in 1972. His approach to preparing for government was revolutionary, entailing party reform,policy innovation and institutional revitalization. His term, exhilarating atits best, and devastating in its failures, was to provide both resources for,and warnings to, subsequent administrations, particularly (but not only)Labor governments.

Whitlam, a brilliant opposition leader between 1967 and 1972, droveparty reform and the generation of policy in an extraordinary manner. Hehad been developing his ‘programme’ consistently since the 1950s, and had aclear agenda when he gained the leadership. He was the progenitor of apolicy explosion that made the Labor Party a credible alternative govern-ment after 23 years in opposition. He was, in his view, an initiator: ‘I don’tthink programmes can be devised by the total body. There are some peoplewho can give initiatives. The party platform is studded with things I wasthe first to suggest’ (Walter, 1980: 31).

Though he provided the initiatives, he astutely garnered advice from otherstowards his ends. He gathered a loyal and pro-active staff, and together theylooked to extra-political experts for the stuff of their programme. Whitlamrelied on these staff to do the groundwork and develop the contacts needed tofacilitate his ideas – not only for policy, but for reform of the machinery ofgovernment. A staffer, Richard Hall, for instance, brought Dr Peter Wilenski, a

38 Managers or Messiahs?

public servant and academic, into the loop, and it was Wilenski who pre-pared draft administrative arrangements for the ministry, the organizationof prime ministerial staff, a reorganization of the public service and aCabinet committee system based on the Canadian model. Wilenski’s workwas analogous to that of Richard Neustadt with his advice to John Kennedyon organizing the transition – and Wilenski became Whitlam’s first principalprivate secretary in office (Walter, 1986: 69–75).

Whitlam was explicit about his modus operandi:I think [my best quality] is probably that I will try to work out means ofachieving what I think are my Party’s objectives – how to marshal theexamples in other countries, how to marshal the advisers in this country– and then try to work out how to explain it to people. I think that’sprobably what I’ve done best … It’s what I tried to do.

Some of those who worked with him, however, became aware of a tellinggap: ‘Whitlam lacks a sense of systems – things being linked together inways that make it impossible to change one part without changing the lot. I don’t think he has that sense even about society’. Or, as another sug-gested, ‘The programme was essentially an accretion of detail, policies andbits and pieces. One of the things wrong with the Whitlam government wasthat while he was flying around on all these fronts, there was never a coherentstrategy behind it all’ (Walter, 1980: 33). Nevertheless, the sheer proliferationof policy made Whitlam a winner. First, it guaranteed him a smooth passagewithin the Party, since no one else had the knowledge to contest his argu-ments. Second, against the run-down, uninspired Coalition governmentWhitlam’s initiatives made Labor look as if it had the answers. Third, therewas no one on the government side who could match him effectively in parliament.

Coming to power with a comprehensive programme of answers to the chiefpolicy challenges of the day, Whitlam engineered a transition such as hadnever been seen before, or since. He established a two man ministry with hisdeputy, Lance Barnard, sweeping into action before his full ministry could besworn in. Whitlam adduced arguments for this unparalleled interim arrange-ment and set to work with a will – even the staid Sydney Morning Herald tookto running a panel on its front page listing ‘What the Government did today’.It was a heady start, but feathers were ruffled, not only those of the defeatedCoalition administration, startled by the speed of the transition, but also of senior Labor MPs, who felt the interim ministry should have includedmore, and of backbench colleagues who were simply doubtful. ‘Some of us inCaucus immediately started looking for ways to protect ourselves from whatGough and Lance might do’ (Walter, 1980: 39).

Consistent with this, Whitlam proved too insistently active to be a goodleader of Caucus. He was incapable of the patience and impartiality needed

James Walter 39

to run meetings, wanting to be central to debate. In 1972, Caucus changedits rules, allowing Whitlam to be the first leader in the party’s history notto chair Caucus. He was too concerned with getting on with ‘the programme’to allow others to debate its details. Whitlam often eschewed negotiationand compromise, involving consultation, by forcing issues in Caucus in theguise of a vote on overriding considerations (his leadership, his ministry)rather than simply as votes on matters of dispute. In the electorate, thisapproach led as early as 1973 to accusations of ‘government by crisis’.Caucus itself remained ambivalent: on one hand, many were impressed byWhitlam’s superb parliamentary performance, on the other they resentedhis steamroller tactics and soon tired of having party votes represented asmatters of confidence rather than of policy. Consequently Caucus wouldoccasionally arrest Whitlam’s progress, less from disagreement than as ameans of asserting its rights. A reform PM needs the support of his allies:such checks did not facilitate the transition.

There were early failures of coordination and consultation, for example,Attorney General Lionel Murphy’s spectacularly ill-judged decision to utilizeCommonwealth Police in a search of the Australian Security IntelligenceOrganization (ASIO) for evidence of Croatian terrorist activity, which ASIOhad allegedly failed to provide on request. This debacle set the fashion, butother ministers failed to learn its lesson, asserting their prerogative to act inde-pendently. Whitlam, with neither the attributes of nor the wish to be a teamcaptain, was unable to keep them in line. These were only the first manifest-ations of what was to become generally apparent in Cabinet: that Whitlamwas a bad manager.

Following the Wilenski plan, Cabinet work was systematized, with some devolution of powers to committees, but this accentuated Whitlam’s centrality:

With the profusion of decision-making units in the Cabinet system, thefull Cabinet receded in prominence and Mr Whitlam augmented hisauthority. With several lines of activity going simultaneously, Mr Whitlamwas the only agent able to keep a hand on all the levers. The Cabinet functioned in his image as an instrument for making quick and attractivedecisions (Lloyd and Reid, 1974: 141).

While Whitlam could easily hold sway, Cabinet proceedings were often frag-mented and the volume of work tackled sometimes led to deterioration inquality.

From the first, Whitlam pushed things through Cabinet too fast. ‘Whitlamwanted to have the biggest legislative programme, the most bills, the mostreforms …,’ said one minister; ‘there was no let up from the pressure foraction’ commented another. One year in, another minister remarked: ‘Wehave taken the attitude that commitments in the policy speech delivered by


Mr Whitlam have to be accomplished in 12 months. I see it as a plan ratherthan a timetable. The Government may be going a little too far too quickly’.The first year set the pattern: by 1975, the Parliamentary Counsel was to com-plain that his office was faced with too much legislation to draft: ‘We do ourbest to maintain a standard, but there is far too much …’ (Walter, 1980: 53).

Whitlam’s turbulent transition was all too indicative of what was tocome. The frantic pace had been set by the historic two-man ministry inthose first weeks when it seemed easier to change the world with a flow oflong-pondered decisions, but as the problems of government became morecomplex, Whitlam’s relentless emphasis on adding to ‘the list’ of achieve-ments, became a problem for his colleagues. Most, lacking Whitlam’s abilitiesand his immense appetite for work, were forced back onto their own areas,because that was all they could keep up with. Few had time for the longview. His administration broke records for decisions made and legislationintroduced, but the pressure of business allowed items through that in bettercircumstances might have been more closely scrutinized. One minister,asked if he felt Cabinet ministers could maintain a fair sense of what wasgoing on, replied: ‘Oh, I suppose we did, but they were coming too fast – itwas like going in a flying machine’ (Walter, 1980: 54). The Whitlam gov-ernment effected significant reforms, but too little of his programme wasconsolidated – in part because changing economic circumstances under-mined the reform project, but in part because one of the most brilliantopposition leaders proved incapable of managing a transition to power thatwould lead to sustainable government. Convinced of his course, confidentof his destiny, Whitlam would fatally misjudge circumstances – the story ofhow he was outmanoeuvred and of his government’s unparalleled fallneeds no recapitulation here (see Kelly, 1983), but its seeds can be seen inthe transition itself.

Whitlam, convinced, as were many of his followers, that his loss of gov-ernment in 1975 was a result of political betrayal (Whitlam, 1979) and thatthe people would see the light, did not resign until another federal electionloss in 1977 showed the futility of that course. He was then succeeded byBill Hayden. Following Whitlam’s resignation Labor worked hard to developa new policy agenda, founded on economic responsibility (foreshadowedby Hayden’s 1975 budget as Whitlam’s last treasurer). A cohort of new, talented backbenchers, many elected in 1977, devised a fresh approach tosocial democracy under Hayden’s leadership, gaining momentum after the1980 elections, which saw a swing to Labor, though not enough to unseatthe conservative Coalition. Intent on avoiding the poor organization andineffective decision-making of the Whitlam years, they prepared carefullyfor government not only through comprehensive policy development but also with successive working parties and task-force reports on cabinet,committee systems and public service reform which issued in a campaigndocument, Labor and quality of government (cited in Wilenski, 1986: 269).

James Walter 41

Peter Wilenski, again, was an advisor in this process. Although Whitlamtoo had planned for administrative reform, this was the first time a partydeveloped transparent suggestions about the machinery of government aspart of its appeal to the electorate.

There was, hence, a well-developed transition plan (Weller, 1983), andHayden could justifiably claim that, ‘the policies on which Labor foughtthe 1983 election had been elaborated and refined … during the previousfour and a half years of my leadership’ (Hayden, 1996: 375). Yet Haydenwould not lead the 1983 campaign. Doubts assailed his colleagues aboutHayden’s ability to best Liberal prime minister Malcolm Fraser, doubts inpart fuelled by the challenger in the wings, Bob Hawke. Hawke, a unionleader and a significant player in Labor politics, was one of the best-knownand most popular figures in Australian public life at the time. There hadlong been talk of drafting Hawke into parliament, and when he won pre-selection for Wills in 1979, and parliamentary election in 1980, his sup-posedly imminent ascension to the leadership was regarded as the fulfilmentof manifest destiny (not least, by Hawke).

That ascension was not smooth: Hawke struggled to adapt to his newmilieu of parliament, while Hayden insisted that Labor’s policy develop-ment, alongside the increasingly apparent differences in the Liberal Party,would ensure victory (Kelly, 1984). The internal party battle was finallyresolved in February 1983 when Hayden decided to call for a vote on theleadership, which Hawke won – catching out Fraser who had called an earlyelection on that same day to try to capitalize on the ALP’s leadership turmoil.Fraser, wrong-footed at the outset, never regained the initiative and lost theelection. Hawke served as Labor prime minister from 1983 to 1991. The deci-sive factor in Hawke’s favour, in the 1983 campaign and long after, was hiscolleagues’ acceptance that his unique public popularity would guarantee the party’s electoral success. In the 1983 election Hawke ran ‘not as leader ofthe opposition but as prime minister-in-waiting’ (Freudenberg, 2003: 89).

The transition plan provided well for the mechanics of government, thoughthere would be successive adjustments as it was implemented, but the socialdemocratic policy agenda would become subsidiary. Believing ‘if the worlddoesn’t trust you then it can ruin you’, Hawke established, ‘responsible’ econ-omic reform as his overriding objective. He allowed others in his ministryscope to defend social goals while he and his treasurer, Paul Keating, busiedthemselves with the economy. They repositioned Labor as the party of marketoriented, realist government, pursuing private sector deregulation, publicsector reform, and economic growth. An Accord with the unions, contain-ing wage claims in return for reinforcement of the social wage, underwroteHawke’s economic achievements. The Accord, a substantial element withinthe transition policy portfolio, was developed by Ralph Willis (shadow trea-surer in opposition) in cooperation with union leader Bill Kelty, rather thaneither Hawke or Keating (the new treasurer).


Hawke’s success depended on his ability to sustain electoral popularity, and the way he managed government. An early study suggested that Hawkeadopted, from the start, a hands-off, almost laissez faire approach to his min-isterial colleagues, that he was not an ideas man, and had no agenda and littlevision (Hede and Wear, 1995). This was to be a recurrent criticism (Anson,1991), yet it ignored the strengths that made his one of the most success-ful transitions to government, and later one of the most successful admin-istrations, in the post-war period. Transition planning clarified the respectiveroles of cabinet and caucus, and allowed for an acceptable split between the inner cabinet and the ministry at large, an effective committee system,and the institutionalization of factional input in a way that facilitated con-sensus. Hawke can take little credit for this, but could capitalize on it withflair. His greatest strength, however, was that he gave a talented team the lati-tude to get on with their jobs. He was an adept negotiator and an excellentchair.

His skill was to focus the cabinet on essentials and to keep it there … Hawke was concerned as much, and at times more, with the processof decision-making as he was with particular outcomes … concerned to conciliate, to identify the points of difference and then to reach an agreement … He acted as chairman, publicist, trouble-shooter and, at times, guardian of the central strategy of government … He was per-sonally involved in detail in only a few areas of policy (Weller, 2007:171, 172, 175).

The achievement of compromises, where each interest gained some (butnot all) of its demands, was Hawke’s speciality. If the Accord was the highpoint, a sequence of Hawke initiated national summits early in the transi-tion aimed at the same result. These were also, partly, an educative process– peak bodies came together to talk, the media reported their exchanges aspart of the story of necessary change. These tactics were effective in achiev-ing gains in health, social policy, community services, immigration andmulticulturalism, education – even while the main game remained theeconomy. Winning the party over to his economic agenda hinged less onclever politics than on Hawke’s belief, which many shared, that the ALPcould not continue in government without him: this gave him leverage ininsisting on the politics of pragmatism (Hawke, 1994). He may have con-centrated on decision processes at the expense of ideas, but driving with alight rein and being able to negotiate consensus facilitated a team approachto Cabinet. It allowed hard men at the core of government to run with economic reform, while philosophically committed reformers were givensome latitude to introduce ameliorative measures. And for a long time,unparalleled in federal ALP history, it contributed to the survival of theLabor government.

James Walter 43

When Labor under Keating eventually lost power in 1996, it remained inopposition for 11 years, and endured four leadership changes – by whichtime the need for a fresh approach to transition planning was apparent.The ALP won government again in 2007 with a map for the transition,drawing on a detailed agenda that had been developed in 2004 (ALP, 2004),and refined subsequently by its new leader, Kevin Rudd, and a number of keypolicy wonks in the shadow cabinet. Rudd, elected leader only a year beforethe 2007 campaign, was ‘desperate to come up with a new agenda and to putit out in the media … [he] was to use much of the work that had been com-pleted by [his predecessor] Kim Beazley’ (Stuart, 2007: 247). Nonetheless,Rudd’s pitch to the Party’s national conference in 2007 stressed ‘fresh think-ing’, he undertook a ‘listening tour’ to seek input, and he recalibrated policyin relation to three big challenges, whose successful address was represented as central to future economic prosperity: the rise of China, the education revolution and climate change (Macklin, 2008: 180, 194, 205). After a periodwhen the party seemed ‘beyond belief’ (Button, 2002), the attention both to machinery of government and policy marked a return to the planning that preceded the 1983 ALP victory. And it at last drew electoral support backto Labor.

Anne Tiernan, reviewing Rudd’s first six months in government drewattention to six elements (Tiernan, 2008). First, there was a plan, it drew onthe lessons of the past (including the 1983 transition) and there wereimplementation agents (such as Faulkner) both interested and experiencedin the machinery of government. Second, Rudd’s prior experience as asenior, Labor-appointed, (state) public servant, who had had to deal with abureaucracy long used to responding to a different party, ensured that hemoved swiftly to make contacts, to ensure briefings and to reassert a com-mitment to public service independence and neutrality. Third, Rudd assertedhis authority over the appointment of the ministry, overturning the con-vention that Caucus would elect the ministry (though, since Whitlam, thePM decided portfolios), and moved quickly to put his team in place. Minis-terial staffing entitlements were established, with an open merit process intheir appointment and a code of conduct promising accountability notapparent under former governments. Cabinet was restructured, with Faulknerappointed cabinet secretary (in effect, a roving trouble-shooter), a restruc-turing of committees (to assure a focus on budget, expenditure review andclimate change), and the introduction of community cabinets (to ensurelinks with communities). Fourth, drawing on practices initiated by Whitlam,but further elaborated by every successor, Rudd retained a powerful PrimeMinister’s Office (PMO) and a strong and active Department of Prime Ministerand Cabinet (PM&C) (though with plans radically to reorganize the latter)– a system of advice reflecting his authority as prime minister. Fifth, illus-trating the government’s commitment to contestable policy advice fromdiverse sources, Rudd looked to ‘shake out’ good ideas, drawing on outside


advisors, task forces and inquiries, and promoting a summit, Australia 2020.Sixth, as Tiernan concluded:

By establishing … ‘discipline and effectiveness’, the new governmentwas able to ‘hit the ground running’. It exploited the opportunities of the transition period to achieve some important and symbolic policygoals, and harnessed popularity and good will to frame a new and distinctive narrative of governance (Tiernan, 2008: 70).

On the other hand, the centralization of authority in the prime ministerbecame a matter of concern during the transition. Rudd’s drive and hisextraordinary work habits set the tempo for his team, and were to drive theAustralian Public Service (APS) close to breaking point by May 2008.Tiernan’s positive account of the successful Rudd government transitionacknowledged that his inability to step back might prove a problem. Shealso noted that as PM&C was reconfigured to meet the needs of the newprime minister, there ensued significant instability and turnover within thedepartment (Tiernan, 2008: 71), an instability that seemed to affect thepublic service at large by mid-2008.

Rudd was seen in some respect as the author of all his government did in its first six months: as one journalist remarked, ‘his fingerprints are on everything’. Despite the ALP’s intention of addressing the problem ofconsolidated executive power, some questioned Rudd’s ability to do so: aformer Foreign affairs colleague of Rudd, for instance, predicted he wouldbe a ‘nightmare, an obsessive who would micromanage everything’ (Age, 6 November 2007: 6), and seasoned observers canvassed the indicationsthat he would ‘seize more power as PM’ (Kelly, 2007). In time, the com-mentary on Rudd’s predilection for micro-management, and a controllingcentralization, became more detailed and more persuasive (Grattan, 2008;Murphy, 2008), especially six months or so in, when the honeymoon cameto an end, cracks appeared in the leadership façade, and Rudd defendedhimself by savaging the APS.

Criticism was forestalled, however, as the escalating global financial crisis inSeptember–October 2008 gave Rudd an opportunity to demonstrate decisionand action. Rudd did not wait to follow the lead of comparable countries. Hewas ahead of the field in lobbying Gordon Brown and George W. Bush tomove discussion from the (US and western Europe focused) G7 group of coun-tries to a G20 summit (including the Asia Pacific and Australia). He was readywith a plan to take to the G20 to improve financial regulation, transparency,adequate capitalization, better monitoring (by the IMF) and incentives to promote stability rather than short-term risk. Though Australia was notbeset with the failure of banks, in common with other western polities itmoved quickly to guarantee bank deposits, and to engineer a substantial fiscal stimulus.

James Walter 45

If caution was overtaken by speed and action, Rudd effectively reassured hisaudience with repeated invocations of his government as ‘calm, methodicaland rational in the way in which it approaches decisions’. It was a messagethat voters accepted: there was a substantial surge in support for the govern-ment and its handling of the crisis. Rudd’s image was transformed fromprocess driven technocrat to strong leader, facing unpredictable times calmlyand decisively with an effective team. He could dominate the crisis ‘story’with his depiction of the market failure of ‘extreme capitalism’ justify-ing activist government, and condemning the outmoded ideology of his predecessors.

In the longer term, this question was decisively answered. Rudd’s pen-chant for control, reliance on a closed circle of senior colleagues and PMOadvisors, and imperious disregard of caucus both in reaching contentiousdecisions and abandoning commitments provoked such disquiet that,when the polls turned sour, the party turned on him. In June 2010 he was challenged for the leadership by his deputy, Julia Gillard, and lost,relinquishing the prime ministership within the ALP government’s firstterm. Some however continued to insist that there was deep unhappinessand dysfunction behind the façade. Rudd’s relentless demands were said to fuel high staff attrition rates and ongoing bureaucratic unhappiness(Milne, 2008). The question became whether a successful transition in thefirst 12 months of government, reinforced by a positive public image, couldbecome undermined in the longer term by a dysfunctional controlling anddemanding personal style.

Preparation and transition: Liberal approaches

Malcolm Fraser (Liberal prime minister 1975–83) engineered the circum-stances that led to Whitlam’s dismissal (Kelly, 1983), then presided over abrief caretaker government until an election could be held, and Whitlamwas defeated. He had been party leader for less than a year prior to thiscontroversial ‘constitutional coup’. His months as opposition leader hadbeen devoted to outmanoeuvring Whitlam: his primary avowed task in sofar as it was announced was to stem what he and his supporters saw as theprofligacy and bad management of the Whitlam years. The big spending ofthe reform project was to be curtailed; the institution of sound processes of management, particularly in the operation of cabinet, was a priority.Margaret Thatcher, preparing a visit to Australia in 1976, was said to be fascinated to learn how Fraser had succeeded ‘without having any policy’(Young, 1990: 129).

What was first evident in the transition was how leadership would beexercised. It followed the practices Fraser adopted in former ministerialroles. From the mid-1960s through to the early 1970s he had developed aparticular administrative style: intense application, immense demands on


his staff, advice from personal networks, canvassing a range of opinion,then a personal decision (Weller, 1989). He rapidly instituted patterns ofwork in the transition that were to characterize his government.

Fraser put great effort into formalizing cabinet processes, managing thepaper flow, formulating the agenda and being well-briefed on every itemthat came before it. His was from the outset a government of constant con-sultation, the obverse of Whitlam’s style. He gained respect and con-currence: one study of his colleagues ascribed to him the characteristics of a transformational leader who gained respect, set high standards, com-municated expectations of high performance, demanded identification of key aspects of complex problems, required opinions to be backed bysound reasoning and imparted a sense of mission (Hede and Wear, 1995).His style has been summarized thus:

[Fraser] … knew what was happening in all areas of government; hepushed ministers to consider the options. He left as little as possible to chance, exploring ideas and trying to assess the outcomes. He testedpropositions by interrogating their advocates, trying to understand thebasics of their argument. He … [argued] that an understanding of thedetail was the necessary prerequisite for seeing the broader picture (Weller,1989: 399).

Fraser also consolidated the resources that Whitlam had introduced – amore powerful, coordinating PM&C and a streamlined personal staff (PMO)tailored to giving him multiple options and a monopoly of information.

There was a flip side to this dedication. As the transition progressed itbecame evident that Fraser, himself tremendously hardworking, expectedas much from his colleagues and was not always considerate of them in theway meetings and committees were scheduled, or the volume of businesshe pushed through cabinet. Everything was interrogated, with insufficientdiscrimination between decisions that were important and those that wererelatively minor and could have been dealt with more efficiently outsidecabinet. His interpersonal dealings were awkward: ‘A shy man, who foundit difficult to encourage others with bonhomie and humour, Fraser dealtwith colleagues by being single minded and determined’ (Weller, 1989:401). His constant attention was perceived by some ministers as a failure oftrust, and his single mindedness as confrontational and ruthless. It was adynamic that generated tension (even fear) and that ended in governmentby exhaustion: ‘Too much time, too many papers, too much consultation,too much interference in ministerial or departmental affairs …’ (Weller,2007: 157).

Fraser’s administrative style made him the centre of networks that weredifficult to challenge: he was the repository of the knowledge needed tomake authoritative decisions. But his incapacity to delegate, his urge to be

James Walter 47

fully versed and oversee every policy option also had a paralysing effect.The weight of detail and excess of knowledge became an enemy of surejudgement. Belying his reputation for ruthlessness and imperiousness, helacked nerve as a decision-maker. He was also the victim of the brinkman-ship that brought him power – not wanting to exacerbate the divisionsunleashed by the constitutional crisis of 1975, he felt condemned to con-straint (Ayres, 1987: 304–6). While it was common to represent Fraser as the harbinger of Thatcher and Reagan, within the parliamentary partyhe came to be regarded as insufficiently rigorous. The conventional Partycritique of Fraser came to be that he led a government of lost opportunities(Ayres, 1987: 304; Richardson, 2001).

The Liberal-National Party coalition remained in opposition for 13 yearsfollowing Fraser’s defeat in 1983. There were attempts at reformulating imageand policy to indicate how a return to government would be approached,with a Future Directions package (steered by John Howard) in 1988 and, morerigorously, with then leader John Hewson’s economic programme, Fightback!.Introduced in November 1991, it was the platform for contesting the 1993election. Hewson proved inept at selling this radical free-market programme(Walter, 1996: 70–6), suffering relentless demolition of his position at thehands of then Labor prime minister Paul Keating (Gordon, 1993: ch. 10;Watson, 2002: ch. 9). Hewson lost an election many had thought unlosable,and was politically finished soon thereafter. Eventually, stalwart former partyleader John Howard was redeemed and returned to the helm at a time whenthe long-serving Labor government was well and truly on its last legs. TheLiberals had to think again about a future transition.

Their solution was to focus on values and the government’s failings.Policy discussion proceeded within the leadership, but Howard prepared forthe 1996 campaign by relying ‘on statements of his broad approach to pol-itics to stave off the inevitable press gallery demands for policies. He putthese ideas into a series of speeches … [that] showed voters he really didstand for something, but without the sort of detail that can trip up oppos-itions’ (Errington and van Onselen, 2007: 217–18). Despite very differentcircumstances, this was a reversion to Fraser’s tactic of outmanoeuvringgovernment ‘without having any policy’. The perception was that in 1996the Coalition government ‘eschewed formal transition planning, believingit had cruelled its prospects in the 1993 election campaign’ (Tiernan, 2008:61). It was an election in which the Liberal machine managed image adroitly,but which Labor lost rather than the Coalition won (Williams, 1997).

John Howard (Liberal prime minister 1996–2007) dominated for a decade,but his initial transition was at odds with the surety and discipline he was todevelop as he grew in the role. One observer of that earlier period captures thetenor of commentary at that time: ‘Within 18 months, the Federal Govern-ment has squandered a 45-seat majority. John Howard’s colleagues attributethe blame to him personally, and the criticisms of his leadership in the late


1980s are being revived’ (FJB, 1998: 47). His predecessor, John Hewson, crit-icized Howard’s populism, pragmatism and misjudgement. Key Liberal Partysupporter Lloyd Williams accused the government of being ‘totally out ofstep’ in its approach to Asia. Key expert bodies such as the Productivity Com-mission and the Wallis Inquiry (into the financial system) found fault withthe pace of reform. Widespread business criticisms of weak leadership werecondensed in a stinging rebuke from Rupert Murdoch, ‘who declared Australiawould become a backwater if it failed to implement a radical reform agenda’.In response, Howard’s cabinet was compelled to develop ‘a strategy to counterclaims that it lacked vision and purpose’ (Ward, 1997: 223).

What generated such criticisms? Hewson was reacting to Howard’s failurefirmly to disown the populist anti-immigrant and anti-‘Aboriginal industry’sentiments of Queensland-based populist and political comet Pauline Hanson.Business was provoked by tardiness in introducing reform, reactive deci-sions that appeared to be taken by a few ministers outside cabinet, failureto adopt Wallis Inquiry recommendations, and threats to trade with Asia.To this could be added decisions that provoked the ‘moral middle class’(Brett, 2003) by appearing politically driven: redefinition of election com-mitments differentiating ‘core’ and ‘non-core’ promises; ad hoc public servicereform, with political acolytes installed in key posts; budget cuts and sackingsof departmental secretaries; a sweep of the boards of statutory bodies, stackingthem with coalition allies; swingeing cuts to the ABC; ill-informed criticismof, and funding cuts to, tertiary education providers; precipitate changes to immigration regulations; and failure to address the High Court’s deci-sions concerning Aboriginal land rights. More damaging was Howard’s post-election introduction of a code of ministerial conduct, which fuelled instabilitywhen it led to the demise of several ministers. Howard then effectively disowned this code and redefined responsibility.

Was this, however, as uncertain a transition as it sometimes seemed? First,despite those early stumbles and failures of judgement, Howard always knewwhat he was about, even if commentators initially failed to see it. He may, fortactical reasons, have evaded policy detail in his election campaign, but hisvision was clear regardless. It entailed: radical industrial relations reform,deficit reduction, further public sector modernization (encouraging privatesector provision wherever possible), a conviction that rent-seeking elites dom-inated most programmes and much service provision (and must be challenged by opening up market competition), a commitment to liberating individual‘choice’ as the best driver of resource allocation, and a conviction that he wasin touch with the core values (‘self reliance, a fair go, pulling together andhaving a go’) of the mainstream. These elements were signalled in his earlyspeeches (see, e.g. Howard, 1996; Errington and van Onselen, 2007: 206–38),and he would preface every policy initiative with comments showing how itwas informed by these core principles, though it was not until his secondterm that the message began to sink in. Second, Howard had learned the folly

James Walter 49

of frenetic change from his experience in the Fraser government and as animpassioned critic of Whitlam’s transition circus. His approach was alwaysgoing to be more incremental than theirs, and those who thought at first thatnothing much was happening (e.g. Ramsey, 1996) failed to recognize thatHoward was there for the long haul. Third – and this, above all, was starklyevident in the transition – he had a sure sense that success depended on con-trolling the levers of power. Getting ‘his’ men into the key posts; buildingfurther on the advisory structures developed by his predecessors (Tiernan,2006); and driving the public service further towards responsiveness wereequally as important as new policy directions, and had to be achieved intandem. Fourth, he was determined from the outset to run a ‘proper’ cabinetsystem (Kelly, 2005; Weller, 2007: ch. 10). He would not be as haphazard as Whitlam, nor would he govern by exhaustion as had Fraser, nor would he bury cabinet in business as both had. The ministers who transgressed theministerial code and had to be removed in his first term provoked cabinetshuffles and reconsideration, which prompted him to reinterpret responsibil-ity as responsibility to the prime minister, and reinforced his conviction that adisciplined cabinet was integral to a long-term government.

Unlike Fraser or Whitlam, Howard appeared to learn from the transitionsetbacks how to put his ideas about using the levers of power and runningcabinet properly more effectively into practice. He became the most effec-tive practitioner of cabinet government in 30 years: running ideas thoughthe party room and into cabinet, insisting on cabinet debate, focusing pre-sentations, listening to views but then locking his colleagues into cabinetdeterminations and achieving a dominance that deterred dissidents andleaks (Weller, 2007: ch. 10).

Styles of taking over: Comparison and discussion

There are consistent differences in the way the major parliamentary partieshave managed their transitions to government. The Labor Party favours plan-ning, relying upon programmatic policy development, calculated reform ofthe machinery of government and identified change facilitators. Respond-ing to challenge, the Liberal Party has experimented with such transitionmaps, (Howard’s Future Directions, 1988, Hewson’s Fightback! 1991–93), butthe failure of the latter saw the Party revert to its default position; cautionabout overt planning and reliance upon the quality of leadership, the leader’sability to articulate values, and political tactics to win power ‘without policy’,as Margaret Thatcher curiously noted. Both Fraser and Howard came to powerthis way, but the pattern has a long history (see Brett, 2003: 30–3). Deep-seated predispositions may be an influence: conservatives have always sus-pected planning to be a step towards state socialism; social democrats usuallycommit to systemic change (and Whitlam testified that Labor’s thwarted post-war reconstruction plans were his initial inspiration).


How significant has planning been in successful transitions to govern-ment? Whitlam demonstrated, for instance that fealty to a programme canimpede a government’s ability to adapt to changing circumstances. On theother hand, Howard offered few hostages to fortune, but his governmentappeared bedevilled by uncertainty in the transition and it fostered cynicismby attempting to redefine broad campaign commitments after election to dis-tinguish between ‘core’ and ‘non-core’ promises. Hawke in contrast was giftedwith a transition plan that worked well for the mechanics of government, but he did not hesitate to drop expansionary policies that conflicted with‘responsible’ economics. His success was in how he managed the transition.

In the cases studied here, the decisive factor in how transitions play outproves be the personal style of the leader. The Whitlam plan for governmenthad an enduring influence: its best features would be a template for later ALPplanning, and the institutional reforms it triggered (enabling more powerfulcoordination through PM&C, augmented information resources, public ser-vice responsiveness, enhanced ministerial staff, contestable advice) would be adopted and developed further by subsequent governments of both com-plexions. It was inadequate decision processes and implementation thathobbled Whitlam’s achievement. His gifts as an inspirational leader, a mes-siah who could persuade true believers that they could change the world, werenot fitted for the more mundane tasks of team-building, negotiation, col-lective determination, or adaptability when the ground shifted (Walter, 1980:ch. 8). Howard, in contrast, with no discernible transition plan, was judgedharshly in the transition phase, but came to be seen as the most effective anti-Labor leader of the post-war period. Did he, however, learn from and changeas a result of the transition? What is clear is a dogged consistency in his under-lying tenets and approach; what he learned from the mis-steps of the tran-sition was to hew even closer to his convictions, to control the levers of powerand to exert discipline in cabinet and, through cabinet, over government.

The two most successful leaders in this group in terms of longevity, andof political impact, were Hawke and Howard. Hawke generated adulationbut not inspiration; Howard was a low key but effective political commun-icator; neither of them had the gift of Whitlam to make individuals believethat, as one true believer opined, they could ‘do anything, or be anything’.The first lesson to draw from this is that sound management is a better pre-dictor of survival and success than inspirational leadership. That said, theywere very different managers: Hawke was a negotiator and broker, orientedto finding accommodations between interests; Howard showed all the traitsof a ‘strong-leader’ (Little, 1988), treating impediments as crises that onlydiscipline and authority could overcome (Walter and Strangio, 2007: ch. 1).

Hayden’s plan in 1983 was a gift to Hawke, but Hawke adapted it to econ-omic imperatives, managed a successful transition and developed a formid-able team; the skills exercised then would contribute to the longer-termachievements of his government. Howard’s transition might be the least

James Walter 51

successful of the set, but persuaded him to rely on his core skills and assertan authority that served him well, at least until the habit of power erodedhis once acute political instincts towards the end of his term (Walter,2008). Hawke’s was the most successful government in the period underconsideration.

Despite intense egotism, Hawke did not rely on dominating those withwhom he worked to affirm his importance: he found it instead in his ‘loveaffair with the Australian people’. On one hand, this freed him from the drive for control that preoccupied most of his fellow prime ministers – he could steer with a light rein. On the other, confidence as to his ownindispensability gave him authority and his focus on consensus decision-making facilitated team-building and allowed others to develop policydetail while he set the direction of government to which all would commit.He could understand and analyse a brief superbly, but he would not getlost in detail and information in the way that Fraser did (and perhaps Ruddmight). He was a global thinker who spoke in broad terms of the need for economic reform, but in practice focused on processes rather than out-comes. He would be lambasted as lacking vision, but he was not tied to grandprogramme objectives and could avoid the cul de sac in which Whitlam foundhimself and the rigidity that dogged Howard’s final term.

In 2007, with Rudd Labor again came to power with a plan and it servedwell during the transition: initial assessments were positive and publicsupport remained high. Rudd was not an inspirational leader, despitespeaking effectively to some intelligent symbolic initiatives (the Kyoto pro-tocol, the apology to Indigenous Australians). He neither elaborated core prin-ciples (like Howard), nor a grand narrative (like Whitlam). He appeared (withFraser and Howard) to be towards the controlling and dominating end of thespectrum, attuned to organization but unable to drive with a light rein, or todevelop skills as a negotiator who governs through consensus and develops ateam. Persistent rumours of public effectiveness, but private dysfunction (inrelation to unsustainable demands on personal staff and the APS) signified afailure to recognize organizational limitations. A successful transition was notenough: closeted decision processes compounded political misjudgement toerode his popular support; immersion in detail impeded policy progress; andrefusal to heed caucus built a store of resentment that would be unleashed tobring him down once the tide of opinion turned. Rudd’s spectacular fall pro-vides another instance of the folly of leader dominance that disregards partyrestraint.

This discussion indicates that a bias towards management, team-buildingand accommodation is a better predictor of a successful transition to powerand effective leadership thereafter than inspirational qualities and grandplans. Granted, at some times, a transformational leader, refusing to acceptconventional orthodoxies is a necessity (Roosevelt with his ‘New Deal’ inthe 1930s, for instance). And some circumstances licence an element of


‘craziness’ in a leader (Churchill’s refusal to accept ‘realistic’ assessments ofBritain’s position sustained his country in World War II, for example). Butin general, drawing lessons from both the Labor and the Liberal patterns,we might conclude that over-reliance on either a plan, or solely on a leader,is to be avoided. Transition planning can facilitate the bridge into govern-ment, but the nature of leadership is also decisive. A leader whose authorityis accepted, but exercised with a light rein, drawing on plans that are nottoo rigidly interpreted, and managing talented collaborators, offers the bestprospect for transition success and political longevity.

Bibliography

ALP (2004) Machinery of Government: The Labor Approach (Canberra: Australian LaborParty Policy Document).

Anson, S. (1991) Hawke: An Emotional Life (Ringwood: McPhee Gribble).Ayres, P. (1987) Malcolm Fraser (Richmond: William Heinemann Australia).Brett, J. (2003) Australian Liberals and the Moral Middle Class (Melbourne: Cambridge

University Press).Burke, J. (2000) Presidential Transitions: From Politics to Practice (Boulder, CO: Lynne

Rienner).Button, J. (2002) Beyond Belief: What Future for Labor? (Melbourne: Black Inc.).Errington, W. and van Onselen, P. (2007) John Winston Howard: The Biography

(Melbourne: Melbourne University Press).FJB (1998) ‘Best forgotten’ (Political Notes), Australia and World Affairs, 35, 47–51.Freudenberg, G. (2003) ‘The passionate persuader’, in S. Ryan and T. Bramston

(eds) The Hawke Government: A Critical Retrospective (Melbourne: Pluto Press), pp. 88–97.

Gordon, M. (1993) A Question of Leadership: Paul Keating, Political Fighter (St Lucia:University of Queensland Press).

Grattan, M. (2008) ‘Much to do and to care about’, Age, 14 March.Hawke, B. (1994) The Hawke Memoirs (Melbourne: William Heinemann).Hayden, B. (1996) An Autobiography (Sydney: Angus & Robertson). Hede, A. and Wear, R. (1995) ‘Transformational versus transactional styles of cabinet

leadership in Australian politics’, Australian Journal of Political Science, 30, 469–84.Hennessy, P. (1997) ‘The Blair style of government: An historical perspective and an

interim audit’, Government and Opposition, 33(1), 3–20.Hennessy, P. (2001) The Prime Minister: The Office and Its Holders Since 1945 (London:

Penguin).Howard, J. (1996) ‘The Liberal tradition: The beliefs and values which guide the federal

government’, 1996 Sir Robert Menzies Memorial Lecture.Kelly, P. (1983) The Dismissal: Australia’s Most Sensational Power Struggle – The Dramatic

Fall of Gough Whitlam (Sydney: Angus & Robertson).Kelly, P. (1984) The Hawke Ascendancy: A Definitive Account of its Origins and Climax,

1975–1983 (Sydney: Angus & Robertson).Kelly, P. (2005) Rethinking Australian Governance – The Howard Legacy, Occasional

Paper Series 4/2005 (Canberra: Academy of the Social Sciences in Australia).Kelly, P. (2007) ‘Rudd would seize more power as PM’, The Australian, 7 November.Kelly, P. (2009) The March of Patriots: The Struggle for Modern Australia (Melbourne:

Melbourne University Press).

James Walter 53

Kumar, M. and Sullivan, T. (eds) (2003) White House World: Transitions, Organizationand Office Operations (College Station, TX: A&M University Press).

Little, G. (1988) Strong Leadership: Thatcher, Reagan and An Eminent Person (Melbourne:Oxford University Press).

Lloyd, C. and Reid, G. (1974) Out of the Wilderness: The Return of Labor (Melbourne:Cassell).

Macklin, R. (2008) Kevin Rudd: The Biography (Melbourne: Viking/Penguin).Milne, G. (2008) ‘Rudd rides high but wears out weary staffers’, The Australian,

3 November. Murphy, K. (2008) ‘Rudd’s will to power’, The Age, 29 March.Ramsey, A. (1996) ‘The longest 45 days’, Sydney Morning Herald, 17 April.Richardson, C. (2001) ‘The Fraser years’, in J. Nethercote (ed.) Liberalism and the

Australian Federation (Leichhardt: Federation Press), pp. 214–29.Stuart, N. (2007) Kevin Rudd: An Unauthorised Biography (Carlton: Scribe).Tiernan, A. (2006) ‘Advising Howard: Interpreting changes in advisory and support

structures for the Prime Minister of Australia’, Australian Journal of Political Science,41(3), 309–24.

Tiernan, A. (2008) ‘The Rudd transition’, Papers on Parliament Number 49 (Canberra:Department of the Senate, Parliament House), pp. 59–77.

Walter, J. (1980) The Leader: A Political Biography of Gough Whitlam (St Lucia:University of Queensland Press).

Walter, J. (1986) The Ministers’ Minders: Personal Advisers in National Government(Melbourne: Oxford University Press).

Walter, J. (1996) Tunnel Vision: The Failure of Political Imagination (Sydney: Allen &Unwin).

Walter, J. (2008) ‘Is there a command culture in politics? The Canberra case’, in P. ’t Hart and J. Uhr (eds) Public Leadership: Perspectives and Practices (Canberra:ANU E Press), pp. 189–202.

Walter, J. and Strangio, P. (2007) No, Prime Minister: Reclaiming Politics from Leaders(Sydney: UNSW Press).

Ward, I. (1997) ‘Political chronicle, July–December, 1996: Commonwealth’, AustralianJournal of Politics and History, 43(2), 217–24.

Watson, D. (2002) Recollections of a Bleeding Heart: A Portrait of Paul Keating PM (Sydney:Knopf).

Weller, P. (1983) ‘Transition: Taking over power in 1983’, Australian Journal of PublicAdministration, XLII(3), 303–19.

Weller, P. (1989) Malcolm Fraser PM: A Study in Prime Ministerial Power in Australia(Ringwood: Penguin).

Weller, P. (2007) Cabinet Government in Australia, 1901–2006 (Sydney: UNSW Press).Whitlam, G. (1979) The Truth of the Matter (Ringwood: Allen Lane).Whitlam, G. (1985) The Whitlam Government: 1972–1975 (Ringwood: Penguin).Wilenski, P. (1986) ‘Administrative reform – General principles and the Australian

experience’, Public Administration, 64, 257–76.Williams, P. (1997) The Victory (Sydney: Allen & Unwin).Young, H. (1990) One of Us: A Biography of Margaret Thatcher (London: Pan Books).


4Establishing Prime MinisterialLeadership Style in OppositionWayne Errington

Introduction

Those who make the transition from leading the opposition to leading thegovernment are, by virtue of winning an election, successful leaders. Yet,the early months and years of new governments are often mistake-riddenand directionless. Some new prime ministers, like Britain’s Tony Blair, havelittle or no prior experience in executive government. Blair’s press secre-tary, Alastair Campbell, observed that Blair’s office had ‘hung on to some ofthe techniques and ways of opposition for too long’ and ‘the spin thing’became a bigger issue as a result (BBC News, 9 May 2002). Australian PrimeMinister John Howard, though an experienced minister in a previous gov-ernment and ultimately a successful prime minister, struggled to controlthe issue agenda during his first term from 1996. By contrast, Howard’s suc-cessor, Kevin Rudd, steadfastly refused to compromise the relationship hebuilt with the electorate as opposition leader and reaped political dividendsas a result.

This chapter contrasts the demands of leading the opposition with thoseof leading the government and considers how often the skills to manageboth types of leadership can be found in the same person. Four prime min-isters are compared, two British and two Australian, two Labour (or Labor,in the Australian spelling) – Blair and Rudd – and two conservative (fromthe Liberal Party of Australia and the British Conservative Party) – Howardand Margaret Thatcher. The Howard case study builds upon the extensiveresearch I conducted while writing his biography (Errington and vanOnselen, 2007). The other cases rest on secondary analysis of the subjects’speeches, media reports and interviews as well as studies by other scholars(which were amply available on Thatcher and Blair in particular). The com-parative case analysis is designed to illustrate – not test – a number of keyleadership style variables at play in a transition from opposition to govern-ment. Case examples will be used throughout the chapter to amplify thegeneral argument.

55

Leadership style

Most of the scholarly analysis of political leadership deals with heads of government or cabinet ministers. They are the actors who can transform pol-itics and society. The differences between opposition and government, andthe consequences for governments of the leadership style adopted while in opposition, are less well understood. Where leadership style is a matter of psychology, we would not expect too many differences in managing the transformation from opposition to the treasury benches. Whatever the temperament of a political leader, though, organization and tactical require-ments clearly differ between opposition and government. Governments haveenormous resource advantages over their opponents, making patronage, and therefore stability, easier. Prime ministers are often in a good position to contrast the strength of their decision-making with the weakness and opportunism of their opponents. Most importantly, governments have the advantage of being able to formulate and implement policy, and the dis-advantage of being held responsible when things outside their control gowrong. Taking the idea of leadership style and examining its development in opposition and during the transition to government adds another level ofcomplexity.

The notion of leadership style has its proponents within political science,mainly within its subfield of political psychology. It also has its critics,mostly found within comparative politics. Comparative studies of primeministers tend to focus on differences in the structure and relative power ofthe office. Rose (1991) argues that to the extent that distinct leadershipstyles are discernible, they are governed by institutional structures such as the party system and other variables that allow a greater or lesser cen-tralization of power. Pessimism about the ability of scholars to account forcontingent factors such as leadership is not limited, of course, to the studyof parliamentary democracy.

However, decades of painstaking micro-level research into leadershipstyle have left a rich, multidisciplinary theoretical and empirical literature,with variables as diverse as type of pre-presidential career (executive or legis-lative background) or studies of birth order in the families of prominentleaders (e.g. Valenty and Feldman, 2002). Psychological approaches to thenotion of ‘strong leadership’ types (Bennister, 2008: 335) are popular in theassessment of prime ministers (see also Little, 1988). Leadership style arisesfrom a combination of the leader’s personality and ideology, as well as the circumstances and institutional factors faced by individual leaders. Oras Bennister described it, individual leadership style is subject to ‘location,relation, environment and events’ (2008: 335).

Not all the variables discussed in the literature are of interest to a study ofthe leadership style of parliamentary opposition leaders. One factor that theopposition period can shape (and reveal to the public) is the motivation to

56 Establishing Prime Ministerial Leadership Style in Opposition

lead, something that Kaarbo (1997) has discussed with respect to the foreignpolicy decision-making of prime ministers. Leaders can be motivated by ideo-logy, a desire for public approval or personal gain. Psychological factors, ofcourse, will influence motivation. Yet, opposition forces leaders to reveal pref-erences about the sort of government they aspire to, and the means theyintend to use to achieve it. Richardson et al. (1982) complement the familiartransformational and transactional leadership styles by differentiating betweengoals and methods. Terms such as imposition and consensus, anticipatory andreactive, help us understand the ways in which leaders approach their tasks.However, such typologies assume a goal-oriented leadership style, whereasleaders may be first and foremost office-holders who enjoy wielding powerbut are not motivated by any particular ends.

Presidential style is much discussed in the United States, both amongscholars and the political class (e.g. Greenstein, 2009). By comparison theoffice of the prime minister (as opposed to those who have held it) is a lesspopular subject for style-centred analysis, although scholarship has beenincreasing in recent decades (Walter, 2008). Prime ministers of course learnto lead on the job, or from their direct observations of the job, rather thanimbibing scholarship about the office (King, 1991: 30).

The length of the previous administration is one important variable. For example, replacing long-term governments, neither Blair nor Rudd had experience as cabinet ministers. Their experience of prime ministerialstyle in the parliamentary chamber was provided by leaders of the oppos-ing party. Blair was not shy of admitting that Thatcher provided him withan example of successful leadership. Rudd, too, selected elements of hisstyle based on the success of the man he sought to replace as prime minister.Conversely, Thatcher and Howard both served under leaders whom theywould later criticize. Howard was explicit in attempting to run a more busi-ness-like cabinet than did Malcolm Fraser. While Thatcher may not haveliked the comparison, she clearly learned something about a domineeringleadership style from Edward Heath (King, 1991: 31).

I will argue below that time served and experiences gained as oppositionleader are important, but overlooked, shapers of prime ministerial leader-ship style. Institutional variables play a role here in that Australian par-liamentary leaders are easier to replace than their British counterparts. Blairand Thatcher both led their parties for much longer than Howard andRudd prior to winning an election. An ineffective opposition leader willquickly be replaced in Australia by a vote of the parliamentary party. Whilethis results in considerable leadership instability, it also means that an opposition leader elected a year or so from an election will be greetedwith a voter honeymoon and diminished expectations of a detailed policyplatform (much to the frustration of the prime minister). Indeed, the lastAustralian opposition leader to spend a considerable amount of time in thejob prior to winning an election was Gough Whitlam (1972).

Wayne Errington 57

Opposition leadership as a formative experience

For a new prime minister fresh from an election victory, the tactics developedin opposition will be credited (rightly or wrongly) with the success of theparty in the electoral contest. Moreover, the leader will be credited with theparty’s success. The circular logic of modern campaigning reinforces itself – a combination of media and institutional factors give the party leadersthe bulk of the campaign emphasis, newly elected backbenchers owe theirloyalty to the successful leader, and the leader is in turn given latitude toshape the fortunes of the government as minds swiftly turn to the nextelection. In spite of the obvious differences between the day-to-day require-ments of government and opposition, there are few incentives for a suc-cessful leader to change the way she cultivates her relationship with theelectorate after an election victory.

An opposition has one over-arching goal – winning the next election.Government leaders live immediately with any compromises they make for the purposes of political expediency. In terms of leadership style, fewopposition leaders have the strength to impose policy positions on theirparliamentary team, and while they can play an important role in the par-liament, they are usually making decisions based on the agenda set by the government. Of Richardson et al.’s (1982) policy styles, then, oppositionleaders can be anticipatory or reactive. This dichotomy might be expressedin electoral terms as big target/small target approaches to policy formulation– whether the opposition wants the government or themselves to be fore-most in the minds of voters.

Opposition leaders usually have little choice but to respond to an agendaset by governments. While an alternative government does not want to beseen as purely reactive (Uhr, 2009: 68), they are ultimately judged in com-parison to the government rather than some objective standard of whatopposition should be about. However, leaders who present the electoratewith a modest set of achievable promises can, by establishing a pattern oftrust with voters, establish themselves for a dominant period of govern-ment. The program laid out in opposition may say more about short-termtactics than long-term vision. Still, opposition leaders are preparing theground for governing styles that can feature imposition or consensus. Theymight, for example, seek to contrast their style with that of the incumbent(to be more decisive, in Thatcher’s case, or more humble in Howard’s).

In Westminster parliaments the opposition’s role, as Kaiser (2008: 21)puts it, is primarily ‘to give voters on election day the possibility of makinginformed decisions between the two different political teams and policypackages’. While in recent decades cabinet formation at the provincial levelin Australia has become somewhat more complex with a number of minoritygovernments, the political culture remains adversarial. In Britain, the leaderof the second-largest party is the ‘main opposition leader’, not the only


opposition leader. Further studies of the role of leadership in the transitionfrom opposition to government in systems where multi-party cabinets arethe norm will provide an even greater range of possibilities.

Successful opposition leadership is more than simply a first step along thepath towards achieving goals. Changing leadership styles involves developingthe leader’s relationship with their colleagues and their electorate – electoraltactics can conceivably be disposed of when surrounded by the resources ofgovernment. However, a calculating leader hiding from the electorate the typeof transformation they have in mind is hardly an approach for long-termsuccess. Some type of crisis may prod a leader to change their style, but whilean election victory would instil a leader with greater self-confidence, it wouldnormally only confirm their beliefs in the most effective style of leadership.

We can, of course, cite examples of leaders who promised from oppos-ition a transformation in government. Some, like Thatcher, largely succeeded,though with an important time lag between getting elected and acting uponher underlying ambitions – time needed to assert her authority within cab-inet, party and country. Others, such as the Australian opposition leader fromthe early 1990s, John Hewson, found the electorate a frustration to theirambitions. Whitlam was elected on a programme of reform only to find newobstacles to the fulfilment of his goals. It has become more common in par-liamentary systems for opposition leaders to promise modestly in order toensure that the government’s fortunes are the main target of media exposure.Such leaders still need to justify a change of government. Hence, oppositionleaders appeal to the need for a transformation of personnel – that the gov-ernment is tired, incompetent or unethical (or some combination of these).Ethical transformation was high on the agenda of the leaders dealt with here.Tony Blair benefitted from a series of ministerial scandals under the BritishConservative Party in the 1990s. John Howard promised greater honesty andhigher ministerial standards. Rudd placed a high value on meeting electioncommitments, yet the overwhelming impression was that he got across theline mainly, and merely, because he and his team projected youthful vigour yetlargely promised policy consistency (‘I am a fiscal conservative’, Rudd assuredthe electorate, which doubted his party’s macro-economic credentials).

Opposition leaders are, in a sense, always insisting on transformation. Some,such as Whitlam and Thatcher, taking over from unpopular or ineffectivegovernments, give the transformation message its full effect. Howard, Blairand Rudd, by contrast, were angling for some combination of stability andchange. Given the modern emphasis in modern parliamentary elections on the party leader, the choices made by those leaders affect the balanceeventually found in government between continuity and transformation.

Words and numbers: The tools of opposition leadership

Scholars have long recognized that ‘the confidence-vote that really matters for oppositions is the vote of public confidence’ (Uhr, 2009: 70). Opposition

Wayne Errington 59

leaders, for whom words are the primary (and often only) means to gainpublic confidence, leave a verbal legacy for themselves to deal with in gov-ernment. Unlike multi-party systems, where the electorate is aware of thefact that promises made in opposition are subject to post-election coalitionnegotiations, Westminster systems provide leaders with the opportunityand responsibility to turn rhetoric into action. While the prime minister’smost important contemporary role is that of chief government salesperson,there is the non-trivial business of ensuring that the country is run smoothly,and the dividends brought by sound policy. Government promises will bejudged by the electorate in the context of a record in power. Still, Thatcher’sgovernment faced a good deal of opposition from civil society, a reminderthat prevailing in policy terms goes well beyond winning elections.

Where governments can change the political agenda through their actions,and make their pronouncements sufficiently vague to leave options open,opposition leaders often face demands from the media and from governmentministers to rule certain policies in or out. John Howard found in 1995 thatthe only way he could take a consumption tax off the agenda for the 1996election was to rule it out not just for the next parliament but forever (‘never,ever’ were his words), something he never ever lived down as the ramshacklenature of the existing indirect tax system provoked a rethink in government.Kevin Rudd’s conspicuous promise to end the ‘blame game’, directed pri-marily at federal structures but inevitably applied more broadly, had the effectin government of limiting the extent to which he could credibly blame theprevious government for Australia’s ills after his election in 2007.

Numbers can be just as important as words – particularly the numbers inAustralia’s powerful upper houses. These chambers have been made morelegitimate with the gradual introduction of proportional representation,with governments rarely afforded the opportunity of control of both houses.Opposition leaders, though, use those numbers with care. Powerful upperhouses can be a useful tool for chief ministers in containing the ambitions oftheir more zealous followers. Centrist leaders can blame the failure to imple-ment the most radical elements of the party platform to an unsympatheticsecond chamber.

Opposition leaders can be complicit in this strategy by overplaying theirrole in parliament and not allowing voters to judge the consequences of a government’s true policy preferences. An obstructionist parliament canmore generally complicate the neat lines of responsibility that are (sup-posedly) the chief advantage of a two-party system, with opposition partiesblamed by governments for blocking vital reforms. Whitlam methodicallytook a programme of reform to the 1972 election, claimed a mandate toimplement his policies and called an early double dissolution election in1974 when parts of his programme were blocked in the Senate. The factthat the subsequent joint sitting remains the only instance of a prime min-ister using this constitutional provision to break a deadlock between the


two houses indicates that opposition leaders have been cautious in theiruse of the upper house at Commonwealth level.

The role of opposition in upper houses is further complicated by the factthat the opposition wants primarily to be the government, and therefore likesto be witnessed by the electorate to be finding constructive solutions andcompromises on legislation. The opposition leader therefore treads a linebetween a desire to build credibility as an alternative prime minister and the genuine ideological (or simply partisan) opposition to government policythat motivates followers inside and outside the parliamentary team. It is thiselement of opposition leadership which, notionally, can be easier to managein government, where the leader’s power of patronage ensures that threats are less than idle. However, the leader’s often painful personal experience of division in opposition, more so than any theory of how government shouldwork, can dominate decision-making on how to manage intra-party differ-ences. Rudd’s approach to party discipline as prime minister was to remind hisparty-room about his bond with the electorate, sometimes in rather colourfulterms.

The winning formula

How much transformation can be advocated, then, before it gets too much?Thatcher stands apart among our cases as a leader promising extensive changefrom opposition and delivering much of it, eventually, in government. TheBritish tradition of party manifestos somewhat institutionalizes this approach.Yet, understanding the environment in which this determined leadershipstyle succeeds is crucial. An economic downturn need not bring forth a newapproach to economic management, and not every economic crisis is fatal to an incumbent government. Blair and Howard both learned from earlierattempts by opposition leaders who promised extensive reform from opposi-tion, and were rejected either by the party (as was the case with Howard in1989), or by the electorate, or both. In his first period as opposition leader(from 1985), Howard modelled himself in part on Thatcher. Having been a fairly conventional political figure during the Fraser Government, Howardfound a rationale to lead in promoting new economic ideas within his party,only to find that the Australian public hostile to a programme of privatizationand small government. Howard had ousted Andrew Peacock in 1985 by sug-gesting (from the position of deputy leader and shadow treasurer) that theprevious coalition government, and by implication the then leader, was insuf-ficiently reformist. As leader, Howard took the coalition further to the right,advocating industrial relations reform and privatization. Such claims wereinsufficiently reformist for some conservatives, who on the eve of the 1987election sought a messianic figure in Sir Joh Bjelke-Petersen, the QueenslandPremier – and divided the opposition. Howard’s promise of economic trans-formation was too much for an electorate already weary of the neo-liberalismof the Hawke Labor Government. Many of Howard’s policies would eventually

Wayne Errington 61

be implemented by future governments, but advocating reform proved thewrong approach for opposition on this occasion.

John Hewson’s Fightback! was the full neo-liberal platform. Hewson suc-ceeded in transformation of a sort – initiating a change of prime minister, asan increasingly concerned Labor party ditched its long-time PM Bob Hawke infavour of his more combative former Treasurer Paul Keating – but Fightback!cost the Coalition a supposedly unlosable 1993 election. Hewson provided theanti-model for Howard in terms of both style and substance. Keating’s skills as a political street fighter saw off Hewson and Fightback!. With Howardremoving the sharp ideological edges from coalition policy, Keating’s aggres-sive style would be left exposed. The Goods and Services Tax (GST) pro-vided the largest target in 1993, but Howard was aware that promising toremove the single-payer health insurance system, Medicare, was electoralpoison. Keating’s boredom with adversarial politics, exemplified by his reduc-tion of question time appearances from four to two days a week, only pro-vided more ammunition for the opposition. Once again, Thatcher was not ahelpful model for an Australian opposition with similar policy preferences.

Promising policy transformation from opposition, then, was out offashion in Westminster systems by the mid-1990s. Yet by the same token apotential opposition leader must still convince his party that she will be aneffective leader. The idea of a ‘Third Way’ between socialism and free-marketcapitalism was a convenient label for Blair that navigated neatly betweenScylla and Charybdis. It complemented a political strategy that seemed to fit both the goal-oriented and office-holding models. For the most part, Blairwas choosing not a coherent new platform but an incoherent blend of oldand new ideas, some of them at cross-purposes. One example was crime,where Blair’s mantra of ‘tough on crime, tough on the causes of crime’ cap-tured both the style and the substance of ‘New’ Labour. Blair adopted ele-ments of Conservative populism on crime, an area where rhetoric can oftentrump record in the minds of voters (Downes, 1998: 191). Expensive pro-grammes aimed at prevention were often undermined by the implementationof toughness on crime (p. 192). Yet, the entire New Labour project was acrucial leadership task for Blair, one that marked him in the eyes of the elec-torate as suitable for the top job, and one that gave him an enormous sense ofpurpose. He thus maintained transformational rhetoric at the same time as hedragged the party’s platform to the centre.

Howard, on the other hand, moderated his policy stance while adopting amore transactional leadership style. While Howard maintained the purpose ofa goal-oriented leader, he was able to adapt to the office-holding style for longenough to win in 1996. He provided reassuring rhetoric to interest groupsconcerned about his record, most notably on Asian immigration, both insideand outside the Liberal Party. He ensured colleagues that he would be moreconsultative as leader the second time around. Much to the frustration of theincumbent governments, the positioning of Howard and Blair was enough to


ensure strong election victories. Keating was frustrated that the Howard heknew was not revealed. Similarly, Conservative advertisements hinting at adevilish Blair underneath the moderate clothing proved ineffective.

In addition to the question of just how much and what type of policies to release and when, opposition leaders must respond to the strengths andweaknesses of the leadership style of their opposite number. Having closelyobserved the failure of Mark Latham’s combative style against Howard in2004, Kevin Rudd succeeded in adopting the cautious, small target approachthat Howard had adopted in 1995–96. Indeed, given the age of the govern-ment he faced and the unpopular policies it pursued with a Senate majority,Rudd’s mix of controlled aggression and policy caution contrasted stronglywith predecessors Beazley, Latham and Simon Crean who had too much of either but never both. This seemed just as important to the electorate as his match-up with Howard. Rudd was instantly more popular than any of his predecessors as leader of the opposition. The popular view that he was in many respects a younger version of Howard seemed to do him no harm atall (Adams, 2009).

At the same time, Rudd had the opportunity to distance himself fromHoward on a few important issues such as industrial relations, health andeducation. Labor supporters may have comforted themselves during thecampaign by imagining that their cautious new leader would prove to bemore adventurous in office. Yet, Rudd was quite explicit in claiming a per-sonal mandate as a result of the election victory. He underlined the linkbetween himself and the electorate by ostentatiously choosing his ownministry rather than observing party tradition in allowing the party caucusto elect the ministry. It seems doubtful that Rudd chose many ministerswho would not have been voted in through the factional system, but theprocess was aimed at reminding both the voters and the ministry that itwas the leader who was responsible for the election victory and the leaderwho would be held responsible for the fortunes of the government.

Learning to manage government

How much, though, does political learning in opposition influence a newPM’s leadership style in government? Margaret Thatcher is a good place to start thinking about this. Her ultimate influence on Britain’s politicaleconomy has been debated extensively. The circumstances of her ascensionto power are no less controversial. Perceptions of a national crisis, real orimagined, were important to Thatcher’s electoral success but also to herapproach to leadership from opposition. Thatcher’s combination of strongleadership ambition, ideological goals and top-down (impositional, in termsof Richardson et al., 1982) means could not have been successful (and wouldrarely be attempted) in other democratic political systems. After timid begin-nings she began to dominate cabinet (though never completely): the British

Wayne Errington 63

political system proved flexible enough to accommodate Thatcher’s style ofleadership. The flexibility of cabinet arrangements in Westminster systemsallowed Thatcher and the other leaders in our cases to choose structuresthat suited and reinforced their leadership styles. While structural factorsmight leave more room for centralized leadership in some systems more so than others, it is also important to recognize the role that determinedleaders (or even ineffectual ones) play in shaping convention and traditionin turn (see Rhodes et al., 2009).

Blair brought to his prime ministership a strong sense of where previousLabour governments had gone astray, if not a blueprint for how he wouldmanage his government, and sought to centralize decision-making powerin his office (Hennessy, 2005: 6). This approach was consistent with develop-ments under the Conservative Government. Unlike Blair and Rudd, Howardhad extensive experience of cabinet as treasurer. He envisioned a system thatcombined the efficiency of centralization with the imperative of consultationthat he had learned the hard way in opposition. He also had a strategy to deal with the civil service, sacking a number of department heads. In this area, though, Howard was never confident that he could affect a culturalchange, and sought policy advice from consultants throughout his primeministership.

His bond with the electorate gave Howard – like all successful LiberalParty leaders – the freedom to choose ministers and drive policy develop-ment. He was mindful, though, of the pace of change that the electoratewould accept. Blair’s position was different insofar as there were competingcentres of power within the Labour Party, starting with Chancellor GordonBrown. While Blair was the government’s chief salesman, he shared withBrown ‘policy fiefdoms’ that divided policy development into cabinet sub-committees (Hennessy, 2005: 10). Where Howard wanted cabinet to alerthim to any political perils associated with the government’s position, cabinetfor Blair was largely a way of providing senior government members withinformation about decisions made elsewhere.

Howard’s eventual success as prime minister came as a direct result of thestruggles he faced during his first term. Without the staff turnover forcedupon him by a series of minor scandals, the blend of leadership authorityand consultation that marked his ‘strong leadership’ phase from 2001 wouldnot have been possible. Even though Howard had firm ideas of what cabinetprocesses should look like, he had to learn the best way to put them into prac-tice, and find the best personnel to do the job. He claimed a mandate for hismore controversial policies but the difficulty Australian governments have inpassing legislation through the Senate increases the stakes over this kind ofdebate. It was Howard the compromiser, though, not Howard the policy inno-vator who had won the 1996 election. His promise of virtue in government,buttressed by a tough code of ministerial responsibility, came unstuck when aseries of ministers were forced to resign over technical breaches. Howard


never looked entirely confident until 1998, when he floated a tax reformpackage and squeaked home for another election victory. A renewed policyagenda gave Howard an affirmative argument to make in place of the moredefensive posture that brought him his initial success.

Cutting across the experiences of our subjects’ transitions to governmentwere legacies of their success at media management from opposition. Theability to deliver a consistent, pithy message to the electorate is one thatescapes many leaders of the opposition. It takes discipline – personal andacross the party – and it can be difficult for an opposition leader to benoticed on a day-to-day basis. Media appearances are almost all the elec-torate has to judge the leader of the opposition. It is understandable, then,that one of the chief lessons a new prime minister brings from oppositionis the importance of a consistent media message.

In government, though, media exposure is not difficult to come by.Everything a prime minister does or says will be reported. While disciplineremains crucial and governments are eventually judged by their record, thegovernment’s chief salesman needs something to sell. A policy develop-ment apparatus obsessed with communication may not provide the bestproduct.

Thatcher’s government was at the forefront of emerging techniques in political marketing. Opposition leaders can learn much from observingtheir opponents, but so too do other political actors. New prime ministerscan find that the media and the public quickly tire of political commun-ication strategies that they have already experienced. Further, the centralityof the leader in the eyes of the media can be misleading. Howard could over-rule ministers on policy matters, but his highly structured system of cabinetsubmissions was chiefly aimed at keeping a united front in public, and ensur-ing that the prime minister was adequately briefed when appearing in themedia. All of this reflected Howard’s cautious approach to the business of government as much as any desire on his part to dominate his government.

Communication for Blair was a means of temporarily squaring some ofthe policy inconsistencies that the New Labour project never fully resolved.He could reassure voters about what the Third Way did not mean, but hada much tougher task in explaining what it did mean. Government spokes-people at cross-purposes was an obvious risk. The imperatives of commun-ication were thus built into the policy-making apparatus, designed to enhancethe appearance of policy coherence and a public image of government unityas much as effective delivery. The media’s obsession with Blair’s commun-ication strategy took attention away from the many achievements of his firstyear in office – including devolution, an independent central bank, humanrights legislation and the Northern Ireland peace process.

Due in no small part to the ways in which their leadership unfolded in opposition, Blair developed a governing apparatus in which policy ideasflourished but the mechanics of implementation proved difficult, while

Wayne Errington 65

Howard developed a streamlined system of policy implementation that suffered from the confused ideological position that Howard took into theprime ministership. Where Blair’s style often lacked coherence, Howardoften lacked an agenda in his first term. While in some respects fortunateto have an agenda thrust upon him in the shape of a financial crisis, Ruddentered the prime ministership with a strong sense of how modern govern-ment is managed. The electorate seemed to care little that he lacked thevision of earlier Labor prime ministers.

Rudd’s experience as cabinet secretary in Queensland in the early 1990sgave him the management experience at the core of executive governmentlacking in almost all of his fellow frontbenchers. He was determined to deliveron his albeit modest promises and provide no surprises to the electorate. Ruddvery publicly declared himself an ‘economic conservative’ at his party’s policylaunch during the 2007 campaign. The financial crisis that enveloped theglobal economy 12 months after his election could have given Rudd theperfect pretence with which to abandon many of the conservative economicpolicies he took to the people. Yet, while Rudd has sought to sharpen the ideological differences between the parties in a rhetorical sense, by labellinghis opponents as neo-liberal extremists, he had no intention of abandoningthe (neo-liberal) policy framework of successive Australian governments.

Even before the depths of the financial crisis, some of Rudd’s followerswere advising him to ditch election promises – in particular the tax cutsthat Howard used in a vain attempt to set the agenda of the 2007 election,and which Rudd had largely matched. Sticking with his election com-mitments after the financial crisis unfolded earned Rudd the criticism that ‘political positioning, not good policy, is driving the government’ (TheAustralian, 24 April 2009). Those hoping that Rudd’s donning of the cloakof economic conservatism was only for electoral purposes, and who saw the crisis as a terrible thing to waste, were also disappointed. Rudd did notwaste the crisis – at least not politically. He used it to add crisis manage-ment competence to his list of attributes and maintain dizzyingly highapproval ratings throughout 2009.

The leader’s bond with the public is dependent not just on policies but inpublic perceptions of the character of the leader. Opinion polls indicatethat honesty is the attribute that voters most value in prospective politicalleaders. However quaint this desire for honesty can seem to hard-bittenpolitical observers, Rudd’s steadfast refusal to betray the confidence thatthe electorate showed in him in 2007 was surely the most important factorbehind his remarkably high approval ratings as prime minister.

Criticism of Rudd’s style – his centralized management style, intensemedia management and endless policy reviews – mattered little while hisgovernment was steering away from the unpopular policies of the previousgovernment. The electorate clearly disagreed with the primary criticism ofthe opposition (and much of the media) that Rudd and his government


were ‘all spin and no substance’. When the government ran into trouble in early 2010, it was mainly due to policies developed in response to thefinancial crisis.

The financial crisis was much more damaging to the standing of the BritishGovernment, because the crisis was deeper in the UK and because the gov-ernment had only itself to blame for its predicament. The electorate hadalready responded to David Cameron, opposition leader from 2005, more positively than to any of Cameron’s predecessors who faced Blair. Indeed,Cameron’s strength in the polls brought into sharper focus the question of just what the Conservatives would do in government. Being forced to make difficult choices about spending priorities while in opposition runs the risk that such policies are electorally unpopular, but could prove to beadvantageous once in government.

Learning to speak for the nation

When it comes to policy, prime ministers can do things that opposition leaderscan only talk about. The importance of talk, though, should not be under-estimated. Prime ministers are provided with a unique platform throughwhich to define not just their government and its achievements but thevalues and direction of the nation as a whole. For example, a confront-ational leadership style used to attain electoral victory can in turn damagea leader’s ability to speak for their nation as a whole.

In Thatcher’s case, nationalism was a crucial part of her appeal. Yet thoughshe spoke of common ground, her ability to govern without a plurality ofvotes and free from the constraint of a hostile upper house enabled her to legislate a divisive economic agenda. Thatcher brought to both domestic and foreign policy realms an uncompromising leadership style, in spite of herlack of experience in foreign policy. Nationalism proved an important com-plement to her domestic programme, promising to restore Britain’s greatnessand increasing her popularity with victory in the Falkland Islands war.

Blair’s reputation for political spin quickly saw all of his utterances treatedwith suspicion. Opposition leaders more generally can find it difficult to pivotbetween negative attacks on government policy and the unifying language ofnationalism. While prime ministers, too, can find this balance difficult, theoffice lends a gravitas to the occupant that is denied to the leader of the oppo-sition. Blair’s job in dealing with nationalism was made easier thanks to anopposition bitterly divided over the question of European integration. Yet,like much of the New Labour project, contradictions abounded in Blair’sapproach to nationalism. Labour could claim to be the true national partywhile the Conservatives were largely restricted to representation in England.Blair put forward his devolution policy as way to strengthen the union ratherthan weaken it. In contrast to Australia, though, many of the debates overnationalism centred on ultimately pragmatic matters such as devolution and

Wayne Errington 67

the currency. When it came to rhetoric, Blair was accused of endless plati-tude. Finding the right words after the death of Princess Diana reduced, ifonly briefly, the widespread cynicism towards Blair’s communicationstrategies. On those occasions when a nation turns to its leader for reassur-ance, the image of the cunning politician practised in the art of deceptionproved difficult to shake, particularly during Blair’s divisive attempt to ‘sell’British participation in the American invasion of Iraq.

James Curran writes of Howard’s challenge to Paul Keating’s republicanand cosmopolitan version of Australian nationalism. Howard took ‘everyopportunity possible to impress upon the people his vision of the nation’spast and its contemporary relevance’ (2004: 241). Some of his finest speechesmourned national tragedies or marked wartime anniversaries. A politicianwho was in many respects a divisive figure often rose to the occasion withthe right unifying words. Judith Brett (2003), too, noted the importance ofHoward’s rhetoric in making the Liberal Party relevant to the times. Therewas, though, some defensiveness about the way he went about these tasks.Howard was still campaigning against Keating and his ideas well into hisprime ministership. He seemed to underestimate the extent to which he, asprime minister, had a unique opportunity to speak for the nation, bringinglong-running partisan political arguments to the task. Howard’s response tothe Pauline Hanson phenomenon was influenced by his own treatment onthe national stage when he was accused of racism in 1988. So ingrainedwas the idea in some sections of the Liberal Party that they were undersiege from unnamed proponents of political correctness, that the fact thatthey held the levers of power had little effect on their disposition. In turn,figures such as Howard excused themselves from the responsibility ofletting Hanson and her supporters know when they had over-stepped themark of social respectability. This was an odd position for a conservativeleader to find himself in.

Not all social differences, of course, can be papered over by political rhe-toric. Howard spoke of national unity, but was incapable of practising it. Evenif Howard had included an apology to the ‘stolen generations’ (of Aboriginalchildren forcibly removed from their families into care of the state) amongstthe government’s responses to the 1997 Bringing Them Home report about thetraumas the policies of previous governments had caused, divisive debatesover indigenous affairs would have remained. Such an apology may havetaken the sting out of the lengthy debate over the legislative response to theHigh Court’s 1996 Wik decision about indigenous native title to land, but bythe same token the juxtaposition of events may have made an apology seemrather hollow.

With Howard struggling to find the right balance between partisan andnational sentiment, Rudd was able to occupy a sensible centre ground. Hisdignified apology to indigenous Australians received praise from progressiveslong frustrated under Howard, but Rudd also led the majority of Australians to


a position on the symbolism of indigenous affairs with which they seemedcomfortable. At the same time, Rudd’s no-nonsense leadership style lentitself to adopting large swathes of sometimes tough ‘practical measures’ onindigenous affairs that Howard had said he preferred to symbolic measures.Again, though, the habits of opposition were hard for Rudd to leave behind.On issues of race, he was content to move only marginally from Howard’sconservative position. Sounding rather Blairite, Rudd formulated a position onasylum seekers that purported to be simultaneously tough and humane.Reflecting this, his government subsequently softened some of the hardestedges of Howard’s uncompromising border control regime – a stance thatcame to haunt Rudd when the numbers of illegal asylum seeker boat arrivalsstarted to rise again during the end of his first term. He was confronted with the dilemma of staying the course or doing a U-turn in an area that was keenly exploited by the Liberal opposition and always of great, if sub-conscious, concern to the Australian electorate. All up, Rudd rarely used theunique position of the prime ministership to mark out a new centre groundwhen he had every opportunity to do so during his first term.

Prime ministers can change their staff, their ministers, their cabinetarrangements, their policies, and they can try to change their leadershipstyle. Changing the way that the public views them is a much greater chal-lenge. Demonstrating strong motivation for power can give prime ministersan important platform to discuss issues of national moment. When theyuse that platform, though, the public still sees the partisan politician. Athoughtful speech is no substitute for flag-waving or decisive action. In thisrealm, the goal-oriented leader can succeed where the office-holderflounders.

Governing without opposition leadership experience

Given Bob Hawke’s brief stint as opposition leader in 1983 prior to his record-breaking run as longest-serving Australian Labor prime minister, we mightconclude that little is lost by a political leader gaining the prime minister-ship without carrying the weight of leading a fractious opposition. Yet,Hawke’s ability to fight an election as leader and promise a good deal ofpolicy change marks an important point of difference from those leaderswho arrive in the top job outside a general election. Strongly performinggovernments have little reason to change leaders. Replacing an ineffectiveleader with a new chief minister provides the doubly difficult task of repair-ing the trust of the voters without the authority of an electoral victory. Ofcourse, a governing party may well wish to replace a scandal- or crisis-plagued leader with someone of the office-holding style in order to conveya sense of renewed stability and trust.

An electoral mandate can be a constraint on a political leader in additionto being a powerful political tool. A leader who comes to the premiership

Wayne Errington 69

through a party or parliamentary vote rather than a general election mayhave more freedom of action than a leader who promised much (or little)while in opposition. On the one hand, such leaders generally emerge fromthe senior levels of the government even if, like Paul Keating in 1991, theirchallenge to the leader is made from exile on the back benches. While theirchallenge to the prime minister is an implicit and sometimes explicit repu-diation of government policy, they have to be careful in criticizing decisionsthat they supported as cabinet ministers. As a point of contrast, a leader suchas Nathan Rees, New South Wales Premier from 2008 until 2009 and an MP since only 2007, can present as a fresh face untainted by the actions ofpredecessors, yet perhaps more importantly lack authority with party andelectorate.

On the other hand, prime ministers are difficult to remove from office inthis manner, and a leadership challenge is unlikely to succeed if the exist-ing policy mix has high levels of public support. A new leader may have anopportunity, if not a mandate, to take government policy in a directionthat a longstanding incumbent leader is either unwilling or unable to do.Service in the cabinet may be a blessing in terms of experience of govern-ment, but raises questions about the leader’s motivation. Why does a suc-cessful finance minister need to be prime minister other than for the sakeof occupying the office?

While leadership destabilization is usually about raw political power moreso than philosophical differences, the new leader inevitably requires a publicrationale for the change of personnel. Indeed, in Keating’s case, the reasonfor the challenge was quite literally that it had long been Keating’s turn, sinceHawke had some years earlier promised to step down in favour of Keatingin return for a continuation of the successful political partnership. (A not dis-similar deal between John Howard and Peter Costello over the Liberal Partyleadership emerged publicly in 2006. Costello never challenged Howard, to growing commentariat ridicule). A more polarizing figure than Hawke,Keating proved to be the right leader to take on the ideologically-driven leader-ship of Coalition leader John Hewson in 1993. Hewson’s detailed policy pre-scriptions left Hawke floundering (partly because Keating’s departure for the back benches after a first unsuccessful challenge left the government ranks looking decidedly thin) but the Fightback! blueprint proved a giant bulls eye for Keating. To the extent that a long-term cabinet minister risks beingequally as jaded as a long-term prime minister, some rationale for their leader-ship is required. Keating found his rationale for power in opposing Hewson,through which he was able to articulate a set of values in contrast to those of the opposition. This leadership transition undoubtedly won the LaborGovernment an additional election victory.

Keating and Gordon Brown were both long-term, largely successful financeministers. Circumstances favoured Keating offering new policies, since highunemployment was one of the reasons for Hawke’s declining popularity, and


Keating as prime minister outlined a new programme of public spendingwithout repudiating his own programme of financial deregulation as trea-surer. As Chancellor, Brown had not had a great deal to say about the policythat had most caused Blair to lose public and party-room support – the warsin Afghanistan and Iraq. Yet, Brown proved unable to quickly wrestle his government loose from the commitments Blair had made to these wars. Aslong-serving cabinet ministers, Keating and Brown observed the office of thecontemporary prime minister at close range. Both came to office with firmideas about how the prime ministership should work, and had to deal withthe legacies of their predecessors. It was arguably Keating who was able todevelop a stronger rationale for leading.

Learning to lead in opposition: Conclusions

Leadership style is not immutable. The case of John Howard shows that ashaky transition to power need not mean failure for the administration as awhole. The variables discussed in this chapter are less about the success orfailure of leaders than about the way they go about the business of leader-ship. New prime ministers cannot help but be awed by the records of thosewho previously held the job. There is no shortage of advice, but no how-tomanuals on the transition. While the role of opposition is ‘fundamentallyone of publicity’ (Uhr, 2009: 63), publicity comes easily to governments.Managing the more complex relationship with followers in the electoratethat government demands can be one of the most difficult challenges forprime ministers whose leadership style is usually forged in opposition.

The leaders analysed here managed election victories following a series ofpainful defeats for their respective parties. The cases reveal the interactionbetween time factors over which leaders have little control (length of oppo-sition leadership, which prime ministers they observed first hand), and thechoices leaders make about policy direction and detail. Thatcher had boththe time and motivation to develop policies for which she could claim anelectoral mandate. She proved a poor model for an Australian oppositionwhose opponents were already quite Thatcherite. On his return to the leader-ship, Howard’s motivation to lead was strong, but for both tactical reasonsand the short duration of his opposition leadership his electoral mandate(which was very large in terms of numbers of seats won) was weak. Blairmade good use of his years as opposition leader, but his leadership motiv-ation was about a centrist politics that was effective from opposition butdrew accusations of style over substance in government. Having observedHoward’s strengths and weaknesses, Rudd drew a strong electoral mandatethat served him well during the transition to power, but was less useful as amotivation for leadership over time.

Much government business is not a matter of fulfilling promises in accord-ance with the party platform but dealing with problems as they arise and

Wayne Errington 71

managing the executive on a day-to-day basis. The transformation from oppos-ition to government has been completed when the balance between the com-peting demands of governing and winning elections is found. Blair, Howardand Rudd made varying degrees of adjustment once in office. Others, not for turning, need to get the balance right from opposition.

Bibliography

Bennister, M. (2008) ‘Blair and Howard: Predominant Prime Ministers compared’,Parliamentary Affairs, 61(2), 334–55.

Brett, J. (2003) Australian Liberals and the Moral Middle Class: From Alfred Deakin toJohn Howard (Melbourne: Cambridge University Press).

Curran, J. (2004) The Power of Speech: Australian Prime Ministers Defining the NationalImage (Carlton: Melbourne University Press).

Downes, D. (1998) ‘Toughing it out: From Labour opposition to Labour government’,Policy Studies, 19(3), 191–8.

Errington, W. and van Onselen, P. (2007) John Winston Howard: The Biography(Melbourne: Melbourne University Press).

Greenstein, F. I. (2009) The Presidential Difference, 3rd rev. ed. (Princeton: PrincetonUniversity Press).

Hennessy, P. (2005) ‘Rulers and servants of the state: The Blair style of government1997–2004’, Parliamentary Affairs, 58(1), 6–16.

Kaarbo, J. (1997) ‘Prime Minister leadership styles in foreign policy decision-making:A framework for research’, Political Psychology, 18(3), 553–81.

Kaiser, A. (2008) ‘Parliamentary opposition in Westminster democracies: Britain,Canada, Australia and New Zealand’, Journal of Legislative Studies, 14(1), 20–45.

King, A. (1991) ‘The British prime ministership in the age of the career politician’,West European Politics, 12(2), 25–47.

Little, G. (1988) Strong Leadership: Thatcher, Reagan and an Eminent Person (Melbourne:Oxford University Press).

Rhodes, R. A. W., Wanna, J. and Weller, P. (2009) Comparing Westminster (Oxford:Oxford University Press).

Richardson, J., Gustafsson, G. and Jordan, G. (1982) ‘The concept of policy style’, inJ. Richardson (ed.) Policy Styles in Western Europe (Boston: Allen & Unwin).

Rose, R. (1991) ‘Prime Ministers in parliamentary democracies’, West European Politics,14, 92–124.

Uhr, J. (2009) ‘Parliamentary opposition leadership’, in J. Kane, H. Patapan and P. ’t Hart(eds) Dispersed Democratic Leadership (Oxford: Oxford University Press).

Valenty, L. O. and Feldman, O. (eds) (2002) Political Leadership for the New Century:Personality and Behavior among American Leaders (Westport, Conn.: Praeger).

Walter, J. (2008) ‘Political leaders and the publics’, International Journal of AppliedPsychoanalytic Studies, 5(3), 153–70.


5Bicameralism and the Dynamics ofContested TransitionsJohn Uhr, Stanley Bach and Louis Massicotte

The legislative dimension of transitions

Modern representative democracy is structured around three dimensions orbranches of power: executive, legislative and judicial. Conventional approachesto government transition tend to deal with only one: transitions in and out ofexecutive power. Transitions in general are about new concentrations ofpower but democratic institutional design gives priority to dispersed gov-ernmental powers (Kane et al., 2009). Democracies disperse governmentalpowers across the three branches, which helps explain why incoming gov-ernments are so keen to use their influence whenever opportunities arise to stamp their authority on legislatures (e.g. by claiming ‘mandates’) andthe courts (e.g. by appointing sympathizers). Some constitutional devicesof dispersed power provide even more testing challenges for incoming gov-ernments: federalism, for instance, generates something of a plural exec-utive, nesting incoming national governments in a wider set of politicalexecutives often with considerable power to affect the transition dynamicsof newly elected national executives (Galligan, 2006).

The conventional focus on the executive branch makes good sense becauseoften that is where the main transition game is being played. But not always.The two other branches can throw up obstacles and complexities which delay,modify or curtail smooth transitions expected by incoming executive govern-ments. Many studies of government transition ignore the many ways thattransitions in and out of executive office are affected by the balance of powerin the two other branches. In the limited space of this chapter, we cannot fillthis void entirely. We will say little about the place of the judicial branch instructuring government transitions. Our focus instead is on the legislativedimension of transitions. In particular, we focus on the special challenges that bicameral legislatures may create to ‘regime change’ in the executivebranch.

The term ‘bicameralism’ refers to the system of two-chamber legislativeassemblies, with the term ‘senate’ frequently used as the name of the second

73

chamber. In parliamentary regimes, the first chamber is typically termedthe people’s house or the house of government: two terms that reinforcethe notion that popular government is legitimately determined in the lower house, where the transition between incoming and outgoing govern-ments occurs. The term ‘unicameralism’ refers to the system of single-house legislative assemblies, where all the business of government, including transition, is concentrated in the one house. Thus processes of government transition can differ between unicameral and bicameralsystems.

Bicameralism offers plenty of scope for examples of how legislativebranches can modify the dynamics of transition in ways that rarely attractthe close attention of transition scholars. These modifications do howeverattract the attention of transition advisors in government who appreciatethe potential in bicameral legislatures for passive obstacles to turn intoactive obstruction inimical to the interests of incoming political executives(Uhr, 2006). We accept that bicameral legislatures vary greatly (Pattersonand Mughan, 1999), but our purpose is not so much to delve into those differences but instead to use bicameral cases to illustrate many of theneglected roles of legislative branches in the political dynamics of tran-sition. The chapter covers three bicameral settings: the United States,Canada and Australia. Each case illustrates aspects of what we term ‘con-tested transitions’ arising from competition between holders of executiveand legislative power (Shugart, 2006).

Contested transitions

Our aim is to sketch certain properties of bicameral legislatures that helpexplain the persistence of ‘contested transitions’. The case material we pro-vide describes the institutional environment in which transitions occur. In the limited space available, we concentrate more on the properties of the three institutional settings than on detailed anatomies of specific tran-sitions. Our focus is on the institutional architecture which allows holdersof legislative power to contest an incoming political executive’s transitionto government. Bicameralism provides an important test-case of the capa-city of newly elected governments to ‘take power’ and govern (Tsebelis andMoney, 1997). About three-fifths of contemporary national governments are unicameral, so that in those unicameral political systems that have reg-ular democratic elections we can assume that newly elected governmentshave only one parliamentary chamber or legislative assembly where the new governing party or coalition has to demonstrate its policy and legislativeauthority.

The three examples we have chosen differ in quite significant respects.This allows us to observe variations in the ways that contemporary demo-cracies manage the important business of government transition. All three

74 Bicameralism and the Dynamics of Contested Transitions

nations are federal systems, which is the common explanation for theirshared bicameralism and their ‘Senates’. We emphasize that in each nationthe so-called federal house carries out a range of federal and non-federalfunctions. The US Senate in some important senses modelled the choice for Senates in the Canadian and Australian federations (Uhr, 1998; Smith,2003). All three systems evolved in nationally distinctive ways from the British system of their colonial rulers, with a sequence of half-centuriesseparating the constitutional formation of the US (1787), then the Can-adian (1865), then finally the Australian federal system (1901). The threefederations have very different national legislative assemblies. The US isdistinctive as a presidential system, with separate elections for executiveand legislative branches (Quirk and Binder, 2005). Canada and Australia are parliamentary systems, with the important difference that the CanadianSenate is an unelected chamber while the Australian Senate is elected (bystate-level proportional representation). This rich variation is relevant to our analysis, because our aim is not to devise a simplistic model of transition that can cover as many cases as possible but to supplement and enrich the existing literature on transitions with a more complete picture of the comparative complexities of government transitions.

Not all democratic transitions are alike, and that not even all bicameraltransitions are alike. Many newly elected democratic governments have to learn to share power with non-government political parties, and to learnthat the initial post-election transition to power can remain structurallyand therefore operationally incomplete in many otherwise effective demo-cratic systems. These contested (or negotiated or delayed or bifurcated orincomplete) transitions are not deviations from some supposed democraticnorm, which might be assumed from the neglect of legislative bicameral-ism in the transitions literature. Rather, the three instances we highlightsuggest that democratic transitions have more legitimate variations thanconventionally assumed.

We begin with the United States, not because it is ‘the model’ here butbecause it is the earliest of our three cases to attempt to institute relation-ships between executive government and a federal component of the leg-islative assembly. We then turn to the case of Canada which is a pioneer inblending a federal legislature and a parliamentary system. Part of thatblend is an understandable reluctance to grant too much legitimacy to aSenate precisely because of the potential for such a second chamber toundermine the hard-won rights of ‘responsible governments’ to respect thenorms of parliamentary confidence constitutionally devised in Canada.Finally, we turn to the Australian case which interestingly combines elements of both US and Canadian approaches, with staggered Senate elections creating potential for newly elected governments to face hos-tile Senates denying the government ‘transit rights’ widely expected indemocratic transitions.

John Uhr, Stanley Bach and Louis Massicotte 75

The United States: Keeping the wolves at bay

Every US presidential candidate makes promises he must know he cannotkeep. Any candidate who spoke about the many things that are beyond the control of any president would be castigated by his opponent for show-ing a deplorable lack of leadership. Even before a newly elected president is inaugurated, he already has to be asking himself how he possibly cansatisfy the high expectations that he very deliberately set out to create.Even if he should have been convinced by his own rhetoric, he soon realizesthat the nation’s problems are more numerous, severe, and complicated thanhe had imagined, and that his constitutional power and political influencefor dealing with them are not as great as he had hoped. But if this did not make life difficult enough for a new president, he is immediately con-fronted by three additional problems that are likely to retard his ability to hit the ground running and achieve even his most cherished legislativegoals. Two of these problems involve the US Senate, as may the third.

Nominations and confirmations

There is no shadow cabinet in the United States. One of a president’s firstand most important challenges is assembling his team of department andagency heads and their subordinates. A president selects not only the secre-tary (i.e. minister) of a cabinet department; he also nominates a far largernumber of under secretaries, assistant secretaries, deputy assistant secre-taries, and so on. In American public administration, there is no such civilservice position as permanent or department secretary; the top layers of national departments and agencies tend to be manned by presidentialappointees. By one estimate, a president can make more than 2000 executivebranch as well as judicial appointments, each of which requires confirmation(approval) by a majority vote of the Senate (Tong, 2008).

But the road from nominations to appointments is often long and arduous,precisely because of the need for Senate confirmation. There are variousreasons that might explain why the Senate tends to delay final action onmany nominations (Quirk and Binder, 2005: 407–31). One is the all-consuming intensity of presidential campaigns which leave little time,energy, and attention for such mundane post-election matters as personneldecisions. A second is the amount of information that prospective nom-inees are expected to compile and hand over to presidential assistants and advisors, not only to evaluate a person’s credentials but also to look for potential landmines that could de-rail a nomination in the Senate andembarrass the president. A third is the additional paperwork that Senatecommittees demand of nominees, to assure Senators that nothing deter-minative is being hidden from them and to protect them from confirm-ing a nominee about whom embarrassing or damaging information thenemerges. A fourth is the delicate presidential task of allocating rewards


among his personal supporters and fellow partisans. And still one more is the recent emphasis on demographic diversity in appointments, makingone selection dependent on others and making it more difficult to findpeople who have both the professional credentials and the personal charac-teristics desired for certain positions, and the willingness to endure thenomination and confirmation process.

There is a logic and justification for each of the reasons why it has come totake so long to staff a new presidency. Yet when they all are put together,their collective effect is to severely impede a new presidential administration.Not even the President of the United States gets to pick his own team at will – he has a legislative veto player – the Senate – to persuade, cajole and appeasein a complicated ‘politics of appointments’ (Lewis, 2008).

Divided government

Although presidential candidates often are happy to leave voters with the impression that electing that candidate is all that is needed to reformWashington and solve the nation’s problems, occasionally a candidate will throw an arm, literally or figuratively, around the shoulder of a con-gressional candidate and intone how important it is to send that pros-pective Representative or Senator to Washington to work hand in handwith the presumptive new president. This is about as close as most pres-idential candidates like to come to admitting that there are severe limits onwhat they can hope to accomplish without congressional support.

Yet one of the more striking characteristics of US national politics inrecent decades has been the frequency of divided government, with therebeing a president of one party and one or both houses of Congress withmajorities from the other party (Mayhew, 2005). Between 1901 and the endof World War II, there were only two two-year periods of such divided gov-ernment. Since then, divided government has become common. During the1950s, President Eisenhower spent six of his 12 years in the White Housefacing Democratic Congresses, and since the presidential and congressionalelections of 1968, presidents have confronted a majority of the other partyin one or both houses of Congress in 30 of the 42 years.

Some respected scholars have contended that divided government is notinconsistent with legislative accomplishment and productivity (Mayhew,2005). Yet ask any president or presidential candidate whether he prefers to face a Congress with majorities of his own party, and you may be surethat he will agree. By no means does unified government eliminate theneed for extended negotiation and hard bargaining between the presidentand Congress over the president’s legislative proposals. Nor does unifiedgovernment even assure that presidential legislation eventually will beenacted at all, no matter how much it may be changed during the course of the legislative process. President Clinton was first elected in 1992, forexample, with Democratic majorities in both the House and Senate. Yet the


signature legislative initiative of his first two years in the White House, hishealth care reform initiative, failed to pass. Divided government only makesa difficult process even more difficult, and presidents will have to tread thepath of reform very carefully and smartly (as freshman president BarackObama’s ultimately more successful pursuit of health care reform in 2009–10once again demonstrated).

Changes in American party politics have made the dangers and costs ofdivided government to presidents greater than they had been during muchof the 20th century (Mayhew, 2005: 175–99). The Democratic party enjoyeda string of majorities in the House of Representatives from 1931 to 1995 thatwas broken only twice, in 1947–48 and 1953–54. Yet through almost all ofthis period, Democratic presidents did not enjoy the predictable support of theHouse of Representatives, as a prime minister expects the dependable supportof his parliamentary majority. Internal divisions within the Democratic partymade possible the emergence of a ‘conservative coalition’ of Republican andsouthern Democratic Representatives that often could thwart the legislativeambitions of presidents from Franklin Roosevelt to Jimmy Carter. Divided gov-ernment was a more serious impediment to Democratic presidents, and a lessserious one to Republican presidents, than it has since become.

The American political landscape has changed. In congressional elections,the southern states now vote much more Republican than Democratic; on the other hand, liberal Republicans, especially from the north-easternstates, have almost entirely disappeared from congressional ranks. Eachcongressional party has become more homogenous and the two parties inCongress have become more polarized. More votes now are decided largelyalong party lines. The voting cohesion of congressional parties still doesnot compare with the party discipline that is typical of parliamentary parties,but the difference has narrowed significantly. As a result, it has becomeharder for a president to identify prospective supporters for his legislativepriorities from among the members ‘across the aisle’. It may be easier forpresidents to hold together members of their own party in Congress in sup-port of their legislation, but that is not of very much comfort when theirparty is in the minority in one or both houses.

This is not to suggest that divided government precludes legislative achieve-ment, but it certainly must complicate the life of a new president. The legis-lative process in Washington always involves negotiation and compromise.Only in cases of presidents elected in true partisan landslides – FranklinRoosevelt in 1933 and Lyndon Johnson in 1965 come to mind – can a presi-dent even hope to see his prized bills enacted quickly and without significantchange. In his first years in office, President Reagan was able to transform hispersonal popularity into legislative victories, notwithstanding Democraticmajorities in the House of Representatives throughout his presidency, buteven he had to become skilled at asking for a lot, accepting what he could get,and then declaring victory. The lesson to be learned from Reagan’s experience


is that a president who can provoke fear in members of Congress from theother party – fear that he is so popular among their constituents that theyput their own re-election at risk if they oppose his legislative program – canminimize the additional challenges that divided government pose for anynew president. Few presidents, however, are inaugurated with this advan-tage in hand, and even fewer are able to retain it for very long.

Controlling the Senate?

Finally, any newly elected president had better understand that there is allthe difference in the world between having a majority in the Senate andhaving control of the Senate.

Reporters in Washington often tell their audience that it will require 60 votes for the Senate to pass a certain bill, even though 51 constitutes a majority of the 100-person Senate. This claim is both false and true. It isfalse because, in fact, it requires only a bare majority of 51 for the Senate to pass any bill. However, it also is true because it frequently requires a voteof 60 Senators to enable the body to reach the point of being able to voteon passing a bill by 51 votes.

We tend to associate representative government with majority rule, butthe majority does not necessarily rule in the United States Senate. Perhapsthe body’s most distinguishing feature is the absence in its standing ordersof effective limits on the length of individual speeches or collective debates.Except when the Senate has reached a unanimous agreement to the con-trary, a Senator who has begun to speak may continue to speak for as longas he or she is able. The record remains a 1957 speech against a civil rightsbill that lasted for 24 hours and 18 minutes. Just as important, there is no motion allowed in the Senate to end the debate on a bill, amendment,or motion without thereby also killing it. The result is the filibuster, forwhich the Senate is justly famous or infamous.

Ever since 1917, the Senate has had a rule by which it can force an eventualend to a filibuster by agreeing to a cloture motion, but only by an extra-ordinary majority – that is, some majority larger than a simple majority of 51.In 1975, the Senate adjusted its rules to permit cloture to be invoked by a voteof three-fifths of all Senators – that is, at least 60 of the 100 Senators. And at no time from 1980 until 2009 did either party hold three-fifths of the seatsin the Senate.

The effect is that a large minority of at least 41 Senators, typically fromthe minority party, normally can debate a bill for as long as they wish andthe majority is procedurally powerless to stop them. When this happens,the majority ultimately has only three options: it can attempt to wait outthe bill’s opponents, it can negotiate with them in the hope of makingchanges in the bill that will convince the filibustering Senators to desist, orit can surrender and pull the bill from further consideration. A unified anddetermined minority party in the Senate may not be able to secure passage


of its own favoured legislation, but it usually can block or, at a minimum,seriously delay passage of the majority party’s favourite bills. This has beena reality in the Senate almost from its inception, and one that every newlyelected president understands.

Notwithstanding the procedural power that filibusters give, they wereinfrequent for most of the 20th century, as were motions to stop them. Notso today. There are no satisfactory objective criteria to identify filibusters,but we can note that during the seven decades from 1919 to 1978, a totalof 380 cloture motions were filed; compare this with the 739 motions filedduring the next two decades alone. Correspondingly, twice as many ofthese motions went to a vote during 1989–2008 as during the previousseven decades combined, and two and a half times as many were approved.

Data on cloture motions are an imperfect measure, at best, of the numberof Senate filibusters. Cloture on a bill may be moved when there is no fili-buster or before one begins, and numerous cloture motions may be movedin connection with the same bill and the filibuster on it. There even havebeen cases in which cloture has been moved for reasons unrelated to theprospect or fact of ‘an extended educational debate’. That said, there can be no doubt that the Senate has changed in recent years, and the frequencyof filibusters probably has been the most important manifestation of it. TheSenate today is a much more partisan and considerably less collegial placethan it was in the 1950s, when it was described as a ‘gentlemen’s club’ or asthe ‘world’s most exclusive club’ (White, 1957).

As legislating in the Senate became a much more contested process, bipartisan compromise and cooperation grew more difficult to achieve, as did the legislative challenges to any president, even one whose party enjoyeda Senate majority but not a majority large enough to be ‘filibuster-proof’. Inhis seminal study of presidential power, Neustadt (1991: 10) wrote:

In the early summer of 1952, before the heat of the campaign, PresidentTruman used to contemplate the problems of the general-become-President should Eisenhower win the forthcoming election. ‘He’ll sit here,’Truman would remark (tapping his desk for emphasis) ‘and he’ll say, “Do this! Do that!” And nothing will happen. Poor Ike—it won’t be a bit likethe Army. He’ll find it very frustrating’.

Truman surely exaggerated a bit. Some things a president can accomplish bycommand. For most things, though, he needs the efforts of his appointeeswithin the executive branch and the support of both houses of Congress, andneither comes as easy as any president would like.

Canada: Co-opting and constraining the non-elected

The Canadian Senate embodies a number of conflicting visions (Smith,2003). Like the British House of Lords of earlier days, it was certainly


intended to protect economic elites. In order to fulfil the role of a house ofterritorial representation based on the equality of communities rather thanof individuals, they decided that representation in the Senate would bebased on the equality not of provinces (as in the US Senate) but of three(now four) divisions. The idea of a house of patronage was certainly not farfrom the minds of constitution-makers of 1867, who were well aware thatthe Crown, even then, meant the government of the day (Mackay, 1963).

Two developments that had not been foreseen prior to 1867 soon modifiedthe practical operation of the Senate. First, in contrast with the quick Italian-like succession of cabinets that had been the hallmark of pre-Confederationsettings, the Canadian executive soon evolved into one of the most stableamong British inspired polities. Conservatives dominated the federal sceneuninterruptedly from 1878 to 1896. As for Liberals, their own periods in office(1896–1911, 1921–30, 1935–57, 1963–79, 1980–84, and 1993–2004) were durable enough to earn them the nickname of Canada’s ‘natural governingparty’. As almost all Senate appointments systematically went to friends of theincumbent party, and until 1965 were for life, every new administrationreaching office after a long stay on the opposition benches was bound to befaced with an upper house where the opposition was dominant.

The other development that shook the original blueprint was the progressof democratic ideas. In 1867, there was nothing obscene in granting to anunelected legislative body powers that were equal to those of the electedhouse. Almost all European polities in these days followed the same model.Yet, by mere effluxion of time, it became unacceptable to most that the will of the people be obstructed by unelected legislators, and the Senate found it increasingly difficult to make an effective use of its immense legislativepowers. Hence the conundrum Canadians have had to live with up to now. In theory, the Senate is all-powerful, in practice it is almost harmless. TheSenate soon joined the Crown in the category of ‘dignified’ rather than ‘effec-tive’ bodies. The only way for senators to escape irrelevance was to be seendoing good things: making valuable studies and rectifying the most blatantmistakes committed by the government. The Senate became effectively a houseof review rather than the powerful body that the Constitution suggested(Jackson and Jackson, 2010: 161–6).

In polities where the constitution can be changed more easily, this problemmight have been solved either by making the Senate elective, and moreable to use its powers effectively, or by reducing its formal powers to amore modest and manageable size so that they could be used without cre-ating an uproar. Unfortunately, as is well known, changing the Canadianconstitution has become a nightmare (Jackson and Jackson, 2010: 58-71).For over half a century, Canadians were unable to agree on a formula foramending their own constitution, and the settlement reached in 1982 argu-ably created in this area more problems than it solved. Two attempts at con-stitutional reform since then, in 1987 (the Meech Lake Accord) and in 1992


(the Charlottetown Accord), ended in unmitigated disaster, and constitutionalnegotiations are now commonly equated in Canada with an appointmentwith a dentist. Wholesale constitutional changes often occur when a politysuffers a major trauma, like military defeat or revolution. For better or forworse (on balance: fortunately), Canadians have up to now been spared both.

The context of senatorial transitions

Transitions from one government to the other so far have taken place in acontext of facing a Senate which is usually utterly dominated by their oppo-nents. Upon taking office in 2006, the new Conservative government couldcount on only 23 reliable supporters in the Senate. Liberals had 66 seats, threeseats were held by Progressive Conservatives (PC) who were cool towards thefusion with the Canadian Alliance that had produced the new Conservativeparty. Six seats were held by others, and there were five vacancies. In 1993,Jean Chrétien had faced a Progressive Conservative majority in the Senate.Brian Mulroney’s challenge in 1984 was an even taller order, as Liberals held73 seats in the Senate. In 1957, John Diefenbaker had to compose with aSenate where Liberals had 78 seats against 18 for the PCs and four others.Among contemporary Prime ministers, Lester Pearson in 1963, and Trudeauin 1980, are the only ones who were spared this problem, as Diefenbaker’sand Joe Clark’s respective tenures had been too short to overcome the stronglyentrenched Liberal Senate majority.

Such transitions are bound to be trying. The outcome will depend on twomajor dimensions. How will the Senate react to the new government? Andhow will the prime minister deal with the Senate?

Options for senators

At the beginning of a political transition, Canadian senators face a dilemma.They are widely perceived as a remnant of the rulers Canadians just threw out. Further, notwithstanding the personal worth of many of them, they arecommonly dismissed as the lazy and illegitimate recipients of political pri-vilege. Whether fair or not, the derisive connotations popularly associatedwith senators have arguably made the Canadian Senate one of the most dis-credited legislative bodies in the democratic world. Should Canadians ever be in the position of restructuring their constitution from tabula rasa, there is little doubt the Senate as presently constituted would go (Smith, 2003:149–75; Russell, 2008: 114–15).

All these public perceptions strongly suggest senators had better behavevery carefully when facing a new government, for fear that the public wrathgenerated by their obstruction might eventually lead either to abolition orreform. Senators should work at the margins or do nothing at all thatwould hamper the new government’s agenda, which is after all what theyusually do when their own party is in power. Such an attitude is probablythe instinctive choice of many senators, who are happy with enjoying their


privileges and are afraid to rock the boat by taking themselves, and theirpowers, too seriously.

But senators are also politicians, and among them can be found somefigures who do not believe their own best days have passed yet. They mayargue for a more aggressive stance. This is especially the case when the newgovernment starts challenging established policies and making enemies.Government bills may be rejected or amended in a way that is not accept-able for the government. Borrowing authority bills may be delayed or rejected.The study of controversial legislative measures may also be delayed. Witnessesthe government would prefer not to be heard may be invited to appear beforeSenate committees. Inquiries on issues that will embarrass the governmentcan be launched (Russell, 2008: 112–20).

Senatorial transitions are shaped to a great extent by the attitude of senators,and should they opt for the path of war, they have many weapons at theirdisposal. For anyone looking for an opposition-dominated Senate freelyusing its prerogatives in order to make life difficult for a new government, a prime candidate is the Liberal-dominated Senate in 1984–1990. FromSeptember 1984 to May 1985, the Senate was the only legislative body theLiberals controlled in the country. Prime minister Trudeau had appointedmany from his own inner circle of advisors to the upper house prior toleaving office, so that the Senate could act as a bulwark not only for Liberalpolicies, but for Trudeau’s own brand of Liberalism rather than his suc-cessor’s, John Turner. In succession, the Senate delayed in 1985 a borrow-ing authority bill passed by the House of Commons (with the support ofthe Liberals). The next year, Bill C-67, the ‘gating’ amendments proposedto the Penitentiary Act, had difficulties in the Senate. In 1987, Bill C-22,the Drug Patent Bill and Bill C-84, the Immigration Bill had the same fate.The next year was momentous, as topics of contention between the twohouses included Bill C-60, the Copyright Bill, and Bill C-103, the AtlanticCanada Opportunities Agency Bill. Also in 1988, the Senate rejected theMeech Lake Accord, which had to be passed again by the House of Commons,then went for a showdown by delaying the Free Trade Agreement, leading toan election the Liberals lost. In 1990, they would have rejected the controver-sial Goods and Services Tax (GST), had they not been swamped through theappointment of eight extra senators.

Options for the new government

What now about the prime minister and his government? The chief weaponin the hands of the executive is time. Each vacancy that happens follow-ing a change of government will be filled by a government appointee, andwith years the opposition majority is bound to erode. How much time will beneeded will depend of course on the size of the opposition majority, the ageof opposition senators and their willingness to stay in office. In earlier days,when appointments were for life, as many as nine years were needed before


Mackenzie King’s Liberals, after the 1935 election, could overturn the Con-servative majority in the Senate and build up such a huge majority that six years in office (1957–1963) were not enough for John Diefenbaker to over-come. Even after mandatory retirement at 75 years of age became the rule,eight years would have been needed for Brian Mulroney in order to get amajority in the Senate, had he not chosen to accelerate the process throughswamping. Jean Chrétien had more luck, for three years were enough, but theinitial Tory majority was much smaller.

Apart from smart moves like persuading an opposition senator to retireearlier than scheduled in order to make way for a new appointee (this wasdone under Chrétien), the only way to speed up the work of nature lies in the swamping power provided by section 26 of the Constitution Act,1867. Ironically, earlier constitutional drafts did not include any swampingpower. British colonial officials, aware of precedents in Britain and in itsempire, persuaded the Fathers of Confederation that a limited swampingpower (up to six, then, in a house of 72) was needed. As the provision nowreads (Forsey, 1974), either four or eight extra senators (one or two for eachdivision, so that the interdivisional balance will not be shattered) may beappointed by the Queen, acting on the advice of the prime minister. There isa single precedent for this provision to be used, in 1990 by Brian Mulroney.Once this prerogative is used, however, no new appointments for a divisionmay be made until the representation of that division has fallen back belowits normal size. Swamping in Canada can only speed up the process of change,and cannot overcome an opposition majority over eight.

Meanwhile, prime ministers must bite the bullet and wait. Should theSenate reject a major piece of legislation, they may call an election on theissue and, if successful, hope that the Senate will abide with the conventionthat the people must have the last word on such issues. This is what BrianMulroney did, in 1988, successfully, on free trade with the United States,though Liberal senators quickly found other issues for dispute.

Another option is constitutional reform. Altering the rules of the game withregards to the Senate is difficult. Section 42 of the Constitution Act, 1982, listsfour topics that require the consent of the legislative assemblies of two-thirds(seven) of provinces including more than 50 per cent of the population of all provinces. They are (1) the powers of the Senate; (2) the method of select-ing senators; (3) the number of senators by which a province is entitled to be represented in the Senate; and (4) the residence qualification of senators.Provided the government of the day is able to secure the consent of the Houseof Commons (easy, if you have a majority) and the support of enough pro-vincial governments (not so easy), the Senate is unable to thwart the assaultagainst its composition or prerogatives, for it has in this area a mere sus-pensive six-month veto that the House can overcome. Other dimensions ofSenate reform can be addressed through ordinary legislation, but the lattermust be passed by the Senate as well.


How difficult a modification of the rules of the game can be is illustratedby the unsuccessful attempt by Brian Mulroney to clip the wings of theSenate in 1985 by sponsoring a constitutional amendment that would havecurbed the legislative powers of the Senate. The measure was initially sup-ported by all provinces but Quebec, the latter’s objection having to do notwith the Senate, but with the fulfilment of its own constitutional agenda.Yet, everything soon fell apart in May 1985, when the 42-year old Toryadministration in Ontario lost its majority. It had to give way in July to aLiberal minority government. Without the support of Ontario and Quebec,the consenting provinces included less than half the population of allprovinces, and the measure fell by the wayside.

This leads us to the ongoing transition, generated by the advent in early2006 of a new Conservative minority administration headed by StephenHarper, facing a Liberal-dominated Senate (Jackson and Jackson, 2010: 165–6).The novelty of this transition is the proclaimed goal of the new PrimeMinister to fulfil the old Reform party promise of a Triple E (‘elected, equal,effective’) Senate. The priority given to this goal was highlighted by thetabling in 2006 of two legislative measures. The first one (Bill S-4) reduced theterm of future senators to eight years, though it did not apply to incumbents.The second one (Bill C-43) provided for the holding, within each province,of a consultation of the voters on who would fill a vacancy in the Senate,with the prime minister remaining free, legally, to appoint any other qualifiedperson he wished. How this attempt to tiptoe around the constitutionalobstacles would fare in the courts remains uncertain (Smith, 2009), andthere was little support among the provinces for the change. Meanwhile,Harper emphasized how serious he was about Senate reform by initially refrain-ing from filling any vacancy that arose. There were only two derogations tothis attitude. In early 2006, Harper appointed to the Senate a Quebec cabinetminister who had no seat in the House, in order to improve this vote-richprovince’s representation in his Cabinet. More revealingly, the next year, a Senate vacancy in Alberta was immediately filled by the ‘winner’ of the senatorial consultation held earlier by the provincial government.

By the autumn of 2008, when Parliament was dissolved, none of the two measures dealing with Senate reform had been passed by either house.Further, the Senate, whose consent was needed, was understandably cooltowards both, while some provinces lamented that the federal governmentwas trying to do indirectly, through legislation, what it could not do directly,i.e. through a constitutional amendment needing provincial consent.

Harper was returned in October 2008 with an increased number of seats,but still short of a majority. A provocative financial statement in Decemberresulted in the three opposition parties attempting to defeat the govern-ment and to reach office through a Liberal/NDP coalition supported by theBloc Québécois. The coalition episode ended in failure, but the governmenthad to alter its policies in order to survive. At this point, Harper came to


realize that he had better fill Senate vacancies with his own supporters as soon as he could, which he did in December 2008, August 2009 andJanuary 2010. At the time of writing (April 2010), party standings in theSenate were 51 Conservatives, 49 Liberals, two Progressive Conservatives,and three crossbenchers. It is now only a matter of time before the govern-ment commands a decisive majority in the Senate, which would leave oppos-ition parties in the House of Commons as the chief remaining obstacle toSenate reform.

Harper’s strategy has clearly been effective in keeping the Senate on thedefensive. Because the government was in a minority position anyway,there was little need to stand up and it was wiser to count on oppositionparties in the Commons to keep the Conservative government in check.Even after having filled senatorial vacancies, the government remains com-mitted to reducing the term of senators to eight years, with the proviso thatthis would apply to all senators appointed after 23 October 2008. The Billto that effect was reintroduced in May 2009 and again in March 2010.

Australia: Managing a most powerful ‘house of review’

The Australian story is special in its own way. The formation of Australia as a national political entity followed that of Canada by around a half-centurywhich followed that of the United States by a similar period. Australia is thusthe most modern of the three polities under examination here (Aroney,2009). Australia was in a position to learn from the other two federations and historians have traced the many ways that the Australian federation generation drew upon their British heritage of responsible parliamentary government when searching the American and Canadian cases for insights into modernizing the Australian framework of constitutional government(Alexander and Galligan, 1992).

The Australian constitutional framework shares with the Canadian system a strong tendency to leave many important matters of political manage-ment to convention and the good judgement of elected politicians. Manyimportant operational details relating to elections and the formation ofnew governments are part of the assumed background of the Australianconstitutional system. For instance, the written Constitution has a lot tosay about the Governor-General’s formal exercise of the executive power,but little if anything about the office of head of government, or of politicalparties, or of the timing of elections other than to specify the outer limit ofthe three year term of the House of Representatives and the six year term of the Senate. Remarkably, the written Constitution of 1901 has requiredlittle textual amendment in the century since Federation, suggesting thatAustralian political elites have managed passably well despite the limitedutility of the written text as a guide to the operational realities of Australianparliamentary government.


The constitutional framework identifies the three powers of government:the legislative and executive powers residing in the elected parliament and the judicial power residing in the appointed judiciary (Aroney, 2009: 272–98).The Constitution identifies the Governor-General as the formal holder ofexecutive power but with sufficient clarity that everyone has understood that in almost all circumstances executive power would be exercised by theGovernor-General acting on the advice of an elected ministry. This formalarrangement is important to transitions for several reasons. First, the consti-tutional formalities have allowed the Governor-General to manage transitionsin and out of government in the absence of general elections. This happenedfrequently in the early years after Federation during circumstances of partyfluidity with instances of minor party government. It was also significantduring the early years of World War II when the Curtin ALP governmentcame to power after the ruling conservative coalition lost its majority becauseof a small number of party defectors. And it was at its most significant – andmost controversial – in 1975 when the Whitlam ALP government was sackedby Governor-General Kerr following a protracted standoff in the Senate. Kerrcommissioned the opposition leader as prime minister and head of a caretakergovernment pending an election of both houses, which was subsequentlywon by the caretaker Fraser government with a commanding majority (Bach,2003: 83–119). These instances illustrate the capacity to allow the authorized – constitutionally empowered, yet neutral (Elzinga, 2009) – head-of-state to make important transitions in the office of head of government based on independent judgement about which party or parties enjoy parliamentaryconfidence.

In normal circumstances, transitions between parties holding govern-ment are determined solely by the outcome of general elections. Again,constitutional provisions regulate core aspects of this process: Australianelectoral rules and practice are subject to evolving electoral law throughwhich Parliament has filled out the details consistent with the bare butimportant bones of the constitutional provisions. The most significantissue relates to bicameralism. Although the Constitution does not use suchlanguage, the lower House of Representatives has emerged as ‘the house ofgovernment’ and the Senate has emerged as ‘the house of review’ (Bach,2003: 238–73).

Both houses have virtually equivalent legislative power: the Senate can reject even those few financial bills which it may not amend. Thesmaller Senate has a six year rather than a three year term like the House of Representatives, so that governments are formed on the basis of partyrepresentation in the lower house. Based on the House of Representativessystem of preferential voting in single-member seats, the outcome of gen-eral elections has been remarkably clear for almost all the time since theconsolidation of the two-party contest over government office. Althoughthere have been a number of instances when the ALP has ‘lost the election’


even though its has won a greater share of the two-party preferred votesthan its opponents, the important issue is that all parties acknowledge thatelectoral victory goes to the party winning the greatest number of House ofRepresentatives seats.

The distinctive Australian complication is the Senate. Since the 1949adoption of a system of proportional representation (PR) for Senate elections, governments have not been able to guarantee that their suc-cess in winning control of a majority of lower house seats will be matchedby a majority of Senate seats (Uhr, 1998: 108–15). One explanation is thatvoters in significant numbers split their votes, deliberately complicating the transition to government process by simultaneously casting votes forstrong party government in the House of Representatives and for minorparties as a form of house of review insurance in the Senate. Australia thusstands out as an interesting example of a parliamentary system whereincoming governments can find themselves facing a hostile or even anobstructionist parliamentary majority in the Senate, which has powers virtually equal to those of the House of Representatives. This situation gives Australia an unusual parliamentary version of ‘divided government’where the house of government is held in check by a parliamentary houseof review controlled by non-government interests: sometimes the interestsof the official Opposition party but more often a loose coalition of non-government interests where the balance of power between the parties ofgovernment and the official opposition is held by minor parties and/orindependents.

A second explanation for the discrepancy between control of the twohouses is constitutional. Elections for the House of Representatives occurevery three years or earlier if so determined by the head of government.The Senate however has fixed terms, with half the senators facing electionevery three years (Aroney, 2009: 215–46). Normally, a general election isaccompanied by elections for half the Senate: but the terms of senators arefixed under the Constitution, so that even newly elected senators may nottake up their Senate office until the expiry of the term of the senators theyare replacing. The only exceptions are the four Territory senators estab-lished in 1975 whose terms match those of the lower house. Thus, theearlier that an election is held for the House of Representatives, the longerthe waiting period for newly elected senators to replace those defeated atthe election.

The Constitution anticipates the transition process required by all fixed-term offices, especially those in a parliamentary house that has claims to bein continuous existence on the basis of its rolling or staggered terms. In anysystem of fixed-term offices, there is a requirement that the selection ofincoming senators take place well before the expiry of the terms of thosewhose offices are being contested. Thus section 13 of the Constitution stip-ulates that Senate elections may be held at any time within one year of the


expiry of the relevant Senate terms: which authorizes a transition period of up to 12 months for newly elected senators. The only exception, whichwe will come to shortly, is in the case of a double-dissolution election for all seats of both houses when the beginning of the new term for senatorscoincides with that of the House of Representatives (Bach, 2003: 25–43).

Incoming Australian governments thus can find themselves facing two transition hurdles: first, the lingering presence of Senate representativesof the former governing party whose terms have yet to expire and whoseconstitutional rights block the power of incoming governments to appointany of their newly elected senators to ministerial posts; and second, the prob-able certainty that even when the new senators take up office, the incominggoverning party will not have a Senate majority.

The timing of House of Representatives elections is at the discretion of the prime minister. Only one House has gone its full term (1907–1910),so that we can see the Australian practice is for prime ministers to opt forearly elections, knowing that they have to do their best to articulate thecycles of the two houses. The 2007 transition forms an apt illustration. TheHoward government called the 24 November election on 14 October, dis-solving Parliament which did not resume until after the formation of thenew Rudd government in February 2008. The ALP under Kevin Rudd wongovernment by increasing its number of lower house seats from 60 to 83,compared to the 65 retained by the outgoing Liberal coalition. At the con-current half-Senate elections, the ALP won 18 seats, the same number asthe outgoing coalition. Given the fixed-nature of Senate terms, the Ruddgovernment had to work with ‘the old Senate’ in the five months betweenthe February resumption of Parliament and the inauguration of ‘the newsenators’ on 1 July 2008.

Of particular significance was the balance of party numbers in ‘the old Senate’, where the former coalition parties held a majority as a result of an unprecedented strong Senate showing at the previous 2004 elec-tion. The Howard government enjoyed a bare majority of one in theSenate: the first time any government had a Senate majority since 1980.That government majority of one reverted into an opposition majority ofone after the Howard government lost office in November 2007, therebyfrustrating the ease of the Rudd transition to power. To use the traditionallanguage so well-known by scholars of the Labor party: the party of pro-gressive reform had, once again, won political office without effectivelywinning parliamentary power. The Rudd government had only 28 senatorsduring this transition period, but this rose to 32 with the changeover on 1 July 2008, which also reduced the opposition numbers from that baremajority of one to 37, which took away the official opposition’s clout to stall Senate business, and handed the balance of power back to where it typically resides: in the hands of a loose coalition of minor parties andindependents.


How unusual was the 2007 transition? The knowledge that the Howardgovernment won its rare Senate majority at the 2004 election tells us thatthe 2007 transition was bound to be unusual. It marks the outer limits ofthe transition situation. Most incoming governments face less stressful situ-ations with the Senate. More typical are the 1998, 2001 and 2004 electionswhich, although not leading to any change of government, illustrate thecommon timetable with elections in either October or November, resultingin a transition of at least six months before senators-elect took up office.The 1996 election, which did see a change of government from the KeatingLabor government to the beginnings of the long run of Howard govern-ments, was held in March and illustrates the alternative practice of early-year elections with a much shorter transition period for senators-elect. ThisMarch election timing was the practice in 1990, 1993 and indeed 1983when the Hawke Labor period began.

Three features sum up the Australian situation. First, incoming govern-ments can find themselves unable to deliver key policy promises, such asnew or changed legislation, because of the lingering power of opposingparties in the Senate. In many circumstances, this adversity is only tem-porary and thus stands out as an important example of transition processesthrough which incoming governments frequently have to pass. But passthey do, confident that the old order also passes away. A prominent exam-ple of this policy delay was the frustration evident in the new Rudd govern-ment when engaged in managing public expectations that it would speedilyreform the Howard government’s workplace relations legislation. Not surprisingly, the Senate majority held by the former governing Liberal-National coalition threatened to use its power to hold the government atbay. Only with the change of Opposition leader from Nelson to Turnbull(and later to Abbott) did the coalition parties admit that the Rudd govern-ment probably had earned its mandate to reverse that particular policyframework.

Second, incoming governments can find themselves temporarily deniedaccess to some of their own political talent. The formation of the new gov-ernment includes the composition of a new ministry and a new prime min-ister can face temporary limitations based on the delayed installation inoffice of senators-elect. A test of the importance of this situation would bethe extent to which new governments modified their ministerial makeupby the time of the official installation of newly elected senators. This is asignificant issue given that around one-third of the Australian ministrytends to be drawn from the Senate, regardless of which party is governing.

The third implication is that the frequency of such Senate-related transi-tions accelerates the development of the Senate from a non-governmenthouse to an anti-government house. Perhaps a better description is that theSenate is strengthened as ‘a house of opposition’ where the special con-tribution senators attempt to make is in terms of holding governments to


account (Uhr, 1998: 190–208). The emergence of the Senate from a legis-lative house of review to a house of policy and administrative account-ability is closely related to the transitional dynamics where issues of reviewand accountability first get tested.

Conclusion: Dispersed power spells transition troubles

It is not so surprising that the scholarly literature on democratic transitionshas so little to say about bicameralism, even in federal political systems.Bicameralism is notoriously difficult to treat in comparative terms, witheach bicameral system appearing unique in its own terms (Uhr, 2006). Butit is surprising that so little is said about contested transitions arising fromcompetition over control of legislative power, which occurs quite fre-quently and with greatest visibility in federal systems. We have attemptedto supplement discussion of government transitions by drawing on evid-ence from bicameralism to illustrate the wider relevance of ‘contested tran-sitions’. Bicameralism is simply the means we have selected to clarify the end, which is the challenge incoming political executives face fromcompetition over legislative power.

Part of our aim has been to widen the focus so that transition analystscan take in a greater range of institutional variables when assessing transitionprocesses. Our general view is that attempts to map when power changeshands depends on careful charting of how power changes hands. Ourspecific view is that the how dimension is much broader than simply theassembly of executive talent featured in many conventional accounts ofdemocratic transitions. Our study suggests that three facets of the contestover legislative power should feature prominently in future studies of tran-sitions. First, federalism as a background framework for the formation ofgovernments and for the ‘vertical’ distribution of executive and legislativepowers. Second, bicameralism as a derivative device arising initially fromfederalism but capable of developing its own institutional logic with powerto modify transition plans of executive governments, thereby illustrating a‘horizontal’ separation of legislative powers. And third, the raw power ofpartisan politics in such institutional settings to pose lingering, possiblypersistent, problems for incoming political executives.

Our analysis raises even larger questions about democratic governance.What we describe here as problems for incoming governments are morelikely to be lasting than short-lived. In the United States, even after all thepresident’s executive-office nominees have been confirmed, he may welllack a majority in one or both houses of Congress for part or all of his termin office. The danger of procedural filibusters will remain ever present,barring some revolutionary change in the Senate. In Canada, the stabilityof governments has meant that when there is a change in party control ofgovernment, the new government is likely to find its party in the minority


in the Senate. While it is true that it may be able to reverse that situationeventually by making new Senate appointments of its own party members,that process may take so long that it offers little comfort to the governmentof the day. In Australia, the recent historical record indicates that any gov-ernment, whether newly elected or re-elected, should expect to confront a Senate with a non-government majority, even after the prolonged delaybefore newly elected Senators actually take their seats.

Bibliography

Alexander, M. and Galligan, B. (eds) (1992) Comparative Political Studies: Australia andCanada (Melbourne: Pitman Publishing).

Aroney, N. (2009) The Constitution of a Federal Commonwealth: The Making and Meaningof the Australian Constitution (Cambridge: Cambridge University Press).

Bach, S. (2003) Platypus and Parliament: The Australian Senate in Theory and Practice(Canberra: Department of the Senate).

Elzinga, D. J. (2009) ‘Monarchy, political leadership and democracy: On the importanceof neutral institutions’, in J. Kane, H. Patapan and P. ’t Hart (eds) Dispersed DemocraticLeadership: Origins, Dynamics and Implications (Oxford: Oxford University Press), pp. 119–40.

Forsey, E. A. (1974) ‘Appointment of extra senators under section 26 of the BritishNorth America Act’, in E. A. Forsey (ed.) Freedom and Order: Collected Essays (Toronto:McClelland & Stewart), pp. 50–60.

Galligan, B. (2006) ‘Comparative federalism’, in R. Rhodes, S. Binder and B. Rockman(eds) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press),pp. 261–80.

Jackson, R. J. and Jackson, D. (2010) Canadian Government in Transition, 10th edn(Toronto: Pearson Canada).

Kane, J., Patapan, H. and ’t Hart, P. (eds) (2009) Dispersed Democratic Leadership:Origins, Dynamics and Implications (Oxford: Oxford University Press).

Lewis, D. E. (2008) The Politics of Presidential Appointments (Princeton: PrincetonUniversity Press).

Mackay, R. A. (1963) The Unreformed Senate of Canada, rev. edn (Toronto: McClelland& Stewart).

Mayhew, D. R. (2005) Divided We Govern: Party Control, Lawmaking, and Investigations,1946–2002, 2nd edn (New Haven: Yale University Press).

Neustadt, R. E. (1991) Presidential Power and the Modern Presidents (New York: FreePress).

Patterson, S. and Mughan, A. (eds) (1999) Senates: Bicameralism in the ContemporaryWorld (Ohio: Ohio State University Press).

Quirk, P. and Binder, S. (eds) (2005) The Legislative Branch (Oxford: Oxford UniversityPress).

Russell, P. H. (2008) Two Cheers for Minority Government: The Evolution of CanadianParliamentary Democracy (Toronto: Emond Montgomery Publications).

Shugart, M. (2006) ‘Comparative executive-legislative relations’, in R. Rhodes, S. Binder and B. Rockman (eds) The Oxford Handbook of Political Institutions(Oxford: Oxford University Press), pp. 344–65.

Smith, D. E. (2003) The Canadian Senate in Bicameral Perspective (Toronto: Universityof Toronto Press).


Smith, J. (ed.) (2009) The Democratic Dilemma: Reforming the Canadian Senate (Montrealand Kingston: Queen’s University Kingston, Institute of Intergovernmental Relations).

Tong, L. (2008) The Senate Confirmation Process: A Brief Overview. CRS Report for Congress No. RS20986 (Washington D.C.: Congressional Research Service of the U.S.Library of Congress).

Tsebelis, G. and Money, J. (1997) Bicameralism (New York: Cambridge University Press).Uhr, J. (1998) Deliberative Democracy in Australia: The Changing Place of Parliament

(Melbourne: Cambridge University Press).Uhr, J. (2006) ‘Bicameralism’, in R. Rhodes, S. Binder and B. Rockman (eds) The Oxford

Handbook of Political Institutions (Oxford: Oxford University Press), pp. 474–94.White, W. S. (1957) Citadel: The Story of the U.S. Senate (New York: Harper & Brothers).


6Westminster Norms and CaretakerConventions: Australian and New Zealand Transition DebatesMarian Simms

Caretaker periods and the management of transitions

While research on transitions in Westminster systems is generally agreed to be underdeveloped, interest has been sparked over the past few years withacademics and practitioners discussing caretaker conventions, especially inAustralia and New Zealand. The idea of a caretaker period is central to the ideaof responsible government that is a core Westminster concept. Responsiblegovernment means that the executive is comprised of Members of Parlia-ment (MPs) and that the executive requires the confidence of the Parliament.Consequently when the Parliament has been dissolved for an election theexecutive lacks the capacity to make new laws and the legitimacy to makeexecutive decisions. A neutral public service that continues with routine gov-ernment business during such periods is another hallmark of Westminster.The simplicity of the model, its emphasis on efficient governance, and thelack of detailed prescriptions, meant that it was effective as a means of gov-ernance for a wide range of countries during Britain’s colonial period. Thischapter focuses on the range of interpretations given to the caretaker period.

Terminology is important in these discussions. For some, the term care-taker ‘convention’ is not helpful, or indeed lacks agreed content; and theirperspectives will be outlined below. The term transitions is particularly usefulas it avoids the common pitfall of assuming that there is a set of Westminsternorms or rules which if ascertained and correctly implemented will improvethe quality of democracy. Transitions allow comparative work: for example,studies of the United States presidency indicate both the changing norms overthe ‘midnight’ decisions, and the accompanying media debate. In particularthe type of transition discussed in this chapter is that of the election period;hence it encompasses ‘transitioning out’ as well as ‘transitioning in’. It goeswithout saying that the proper management of transition is a hallmark ofdemocracy.

94

However, the recent country-specific debates have taken different formsand have been generated not only by different issues but by distinct typesof issues. Australia and New Zealand are particularly interesting cases becausethey are similar in some respects (for example as ‘settler colonies’ inOceania) and have an extensive history of policy learning, policy transferand adaptation. In New Zealand, debate has centred on the actions of gov-ernment after elections, whereas in Australia the preoccupation has beenwith the behaviour of government prior to elections. While it is tempting to explain the New Zealand approach as a reflection of the relatively newMixed Member Proportional (MMP) electoral system, this paper argues thatcauses are deeper and with longer historical roots. The introduction of newelectoral systems, such as MMP in New Zealand, or fixed-term parliaments,such as in New South Wales and the Australian Capital Territory, maychange the approach to transition, usually resulting in greater codification.The Australian interest with the pre-election period is particularly intriguing.While a lack of focus on post-election matters could simply be explained bythe election system delivering clear winners, the electoral system does notexplain the discussion over the pre-election period.

This chapter explores the range of factors involved in shaping the dis-course around the transition question in both countries and suggests that amixture of institutional and cultural issues has shaped the debate: electoralsystems have not been the sole or determining factor. The interest in theNew Zealand post-election period initially emerged over the controversysurrounding an outgoing National Party Government’s refusal to take urgentfinancial advice from the incoming Labour Party Government in 1984. Dis-cussions about the political and constitutional ramifications of the pro-tracted process of forming a Government after the first MMP election of1996 (see Boston et al., 1998) drew explicitly upon the lessons from the1984 experience. Whereas in the Australian case, the lack of a history ofwhat might be termed ‘hand over’ or transition politics has led to consider-able discussion over the election campaign period. The furore over the mid-term dismissal of Labor Prime Minister Whitlam by the Governor-Generalin November 1975 may best be seen as the exception that proves the rule.The committing of troops overseas by the Howard Coalition Governmentduring the election campaign of 2001, which sparked widespread commentand further reappraisal, is a case in point (see Warhurst, 2002; Malone,2007; Tiernan, 2008). Malone (2007) notes that in general the debate overcaretaker conventions would be improved by more practical case studies ofcaretaker issues, which this chapter provides.

Westminster conventions: Rules of executive convenience?

Interestingly the New Zealand model for dealing with uncertainty in the post-election period is being promoted for adoption in the United Kingdom (UK)

Marian Simms 95

by the Justice Committee of the House of Commons, as the current UK con-vention deals only with the election period. Amongst the crucial differencesthat have emerged within the Westminster ‘family’ of nations (Rhodes, 2005),cabinet government remains the core similarity (Weller and Keating, 2000)and consequently it is arguable that transitions are central to any discus-sion of contemporary Westminster politics. Trends in such matters may havebecome the only common or shared indicators on the Westminster terrainand provide an important and useful peg on which to hang the discussionover emerging trends in the Westminster family. While flexibility is generallyseen as a key virtue of the cabinet system (Bagehot, 1867/1963), it is also a keydeficiency as its existence is almost entirely governed by convention ratherthan law. According to constitutional expert Wheare (1952), while ‘con-ventions’ are ‘crucial’ to the operation of Presidential systems, as instanced bythe modern interpretation of the electoral college system in the United States,they are do not have ‘constitutive’ roles.

In this chapter, New Zealand and Australia provide the country cases andthe initial focus is upon the significance of electoral systems in generatingpolicy and debate given New Zealand’s introduction of MMP for the 1996election. Generally MMP and similar electoral systems would generate multi-party governance, prima facie incompatible with the parliamentary stabilitynecessary for cabinet governance. A secondary issue, and a closely relatedquestion, is the leadership question. Constitutional expert Ivor Jenningssuggests that the British Westminster system requires strong leaders, espe-cially during times of crisis: leaders who push the boundaries, who are both‘convincing’ and ‘persuasive’ and can hold the system together, includingkeeping ‘in step with public opinion’ (Jennings, 1958: 11). The leadershipissue has at times been problematic for analysts because leadership is deemedby some to be incompatible with democracy. Hence the criticism levelledat Margaret Thatcher’s conservative government as an ‘elective dictator-ship’. A more contemporary take on the topic is to talk about the ‘presiden-tialization’ of responsible government. One may contrast New Zealandwith a parliamentary sovereignty heritage expected to promote strong leader-ship; and Australia where the federal constitution places constraints and chal-lenges upon the prime minister thus requiring bargaining and negotiationskills (Weller, 1985).

Mention should also be made of several other institutions and interestsinvolved in contemporary politics that have helped to shape the debateover cabinet in general and caretaker conventions in particular. In the firstinstance, the media are a ‘driving force’ as a result of the communicationsrevolution and they may set agendas and be utilized by prime ministers and opposition leaders, especially during periods of transition. Secondly, thebureaucracy is often called upon to reshape relationships consequent uponchanges of government. Six permanent secretaries from the Australian PublicService were sacked post-1996 with the election of the Coalition Government.

96 Westminster Norms and Caretaker Conventions

They also have a role in transitions: for example, the Solicitor-General, theCabinet Secretary, the Finance Ministry and State Services Commission allhave parts to play in New Zealand. In Australia, caretaker issues have comeunder scrutiny as part of a broader project, by some bureaucrats to ‘codify’Westminster conventions more generally, and to regulate the behaviour ofMinisters more specifically (see for example Davis et al., 2001; Common-wealth of Australia, 2007; Western Australia, Premier and Cabinet, 2008).Thirdly, political parties may be players. For example, party activists PeterWilenski and Gareth Evans drafted the Australian Labor Party’s very detailedtransition to government document in 1982. This predated the widespreadmove to codify conventions related to transition, and may be seen to haveencouraged this process.

New Zealand’s adoption of the parliamentary sovereignty path (see Wheare,1952) is crucial for explaining much about its approach to transition. NewZealand officials have long drawn upon UK precedents when explaining theNew Zealand approach. The UK history of cabinet governance is replete withexamples of flexibility and outright experimentation, especially in times of war and economic crisis. The core of the so-called Westminster system is‘cabinet government … with which we are familiar in Britain, in other mem-bers of the British Commonwealth and in the countries of Western Europe’;and the main requirement is that ‘the ministers, the heads of the executive,must be members of Parliament’ (Wheare, 1952: 37). The British Parliament is ‘supreme’ in that ‘it was controlled by political means and not by the law of the constitution’ and New Zealand is also in the class of countries without a supreme constitution (Wheare, 1952: 15, 26). In contrast, Australia has awritten as well as a ‘rigid’ constitution in terms of the framework famouslyoutlined by Lord Bryce (see Crisp, 1983).

Two UK examples will indicate the extent of these experiments and theuse of ‘avowed exceptions’ to the rules of cabinet. In May 1915 a CoalitionCabinet of 22 was established under Asquith including 12 Liberals, eightConservatives, Arthur Henderson of the Labour Party, and Lord Kitchener – Chief of the Imperial General Staff. Under Lloyd George in late 1916 theWar Cabinet became the Imperial War Cabinet with the ‘attendance of repre-sentatives of the dominions and India’ and in May 1917 General Smuts of South Africa became a full member of the latter (Thomson, 1965: 48). The ousting of Asquith by Lloyd George in late 1916 conducted in ‘secret,obscure internal processes of which the public only now know the main story’(Churchill, quoted in Thomson, 1965: 47). Then during the Great Depressiona multi-party national Government was formed based on an ‘agreement todiffer’ wherein: ‘Dissentient ministers were allowed to remain within the min-istry: in return for thus submitting to majority wishes, they were given thefreedom to oppose tariffs publicly’ (Thomson, 1965: 144).

The point of this discussion is that Britain, with the doctrine of parlia-mentary supremacy, was able to make such modifications to the ‘norm’ as

Marian Simms 97

outlined by Bagehot amongst others of cabinet as the ‘hyphen’ that linksthe legislature (and hence the people) to the executive (including theMonarch). Transitions were affected as well in the crisis period. Parliament‘surrendered wide powers’ under the Defence of the Realm Acts so that theHouse elected in December 1910 had prolonged its life for eight years; andthe transition from a Liberal to a wide-ranging Coalition Government undera different leader occurred without even a secret sitting of the Parliament.

New Zealand: Rules for effective electoral change

Such UK examples help to explain why so many New Zealand officials dis-tinguish between ‘law’ and ‘practical’ politics and can comfortably leavethe mechanics of transitions to practical politics. For example, former NewZealand Solicitor – General John McGrath – also an important player in the1993 post-election drama (outlined below) – cites A. V. Dicey when main-taining that the term ‘usage’ is a more appropriate word to describe thepoints made in the Cabinet Manual regarding caretaker matters (McGrath,1999). He is supported by leading constitutional lawyers (see Joseph, 2008).

Former NZ Governor General Sir Michael Hardie Boys (1997: 2) also agreedthat transitions were ‘practical’ processes whereby the political parties inthe House make their decisions publicly known and the result is thenimplemented by the Governor-General; and that it has ‘long been accepted… that the incoming administration acts on the advice of the outgoingadministration’. His interpretation has been acknowledged by subsequentGovernors-General. Political scientists tend to seek certainty and perhapssee the Cabinet Manual as a half-way house in the absence of legal or con-stitutional changes, such as fixed-term parliaments, and ‘constructive votesof no confidence’. All are designed to provide greater clarity and to avoidusurpation of power. For others (for example Keating, 1999), the West-minster system of responsible government with its mixture of formal law,conventions and usages functions well enough because of the regular elec-tions that are ensured by key laws, including a written constitution in someWestminster countries, such as Australia and Canada.

Recent work by political scientists on constitutional conventions in NewZealand has appropriately emphasized the impact of the MMP system uponthe conventions of governance, notably the articulating of caretaker con-ventions for the period after an election has been held, when the electionresults are yet to be finalized. In fact a group of writers (Boston et al., 1998)argues that the New Zealand debate in the MMP era overemphasizes thepost-election period, thus shifting the critical caretaker period until after anelection, rather than before and after the election.

While it is doubtless correct that MMP may have influenced Westminsternorms in New Zealand, notably the conventions surrounding collectiveresponsibility (Clark, 2005) and while it is tempting to use MMP as a hook


on which to hang all recent changes in the New Zealand political system, acontrary interpretation is emerging. For example, Raymond Miller (2005)argues that the shape of the party system has been altered less than mighthave been expected. The outlines of an essentially two-party system andcommitment to cabinet government remain. In fact, Helen Clark’s (2005)re-writing of the Cabinet Manual is very relevant here. Her explanation in the ‘Foreword’ to the updated manual states: ‘It amends the doctrine ofcollective responsibility to allow coalition parties to express differentiatedpositions within an agreed framework.’

In respect of the caretaker conventions specifically, there is also agreementthat the political and economic crises of mid-1984, before the implement-ation of MMP, were a catalyst for the formalization of certain caretaker norms.The circumstances are well known and will be summarized only briefly. Anelection called in June 1984 resulted in a landslide against the incumbentNational Party Government of Sir Robert Muldoon. Almost immediately afterthe election the Reserve Bank halted trading, such was the pressure on theNew Zealand dollar. Muldoon, who was Finance Minister as well as the PrimeMinister, ignored the advice from his advisors (senior public servants), frommembers of his own Cabinet and from the Leader of the Opposition (DavidLange) and from the Shadow Finance Minister, to devalue the New Zealanddollar. Finally, after several crisis-ridden days, Muldoon agreed to devalue thedollar. These events have been outlined in a recent biography of Roger Deane(Bassett and Bassett, 2006) who was the Deputy-Governor of the ReserveBank. An interview by the author with the Opposition Whip (5 May 2009)also highlights the role of Muldoon’s Attorney-General Jim McClay in push-ing Muldoon to act on the advice of the incoming government. Boston et al.(1998) also note the similarity between the contents of McClay’s press releaseof the time and the first Report of the Officials Committee on ConstitutionalReform regarding post-election constitutional conventions.

Consequently, a Committee of Officials appointed by the Lange Govern-ment recommended the adoption of written ‘conventions’ relating to thebehaviour of ‘outgoing’ or defeated governments in the post-election period.These will also be summarized briefly: first, that such governments shouldmake no new policies; and, second, that they should accept the advice of theincoming government on urgent matters, even if that advice contradicts theoutgoing government’s policies. These two conventions were subsequentlyaccepted by both major parties and were formalized by publication in theCabinet Office Manual of 1990. This crisis perhaps more importantly led to aconstitutional amendment (1986) with an important redefinition of a min-ister, as someone who had successfully contested the recent election; thusallowing a swearing-in of a new government, before the new parliament hadmet and guaranteed confidence.

The other salient political crisis in the period prior to the introduction of MMP was the considerable delay of nearly two weeks in declaring the

Marian Simms 99

results of the 1993 election, when ‘it was uncertain’ whether the NationalGovernment would be re-elected – in the event it was. The day after theelection, the Prime Minister sought advice from the Cabinet Secretary andthe Solicitor-General to the effect that his government should function in a‘caretaker capacity’ and take only ‘routine decisions’. At the time, this edictwas not constitutionally mandated nor formalized in the cabinet manual – although it was a well established principle in the standard constitutionaltexts. In fact, the 1996 changes to the manual were based on the 1993 pre-cedent rather than simply in anticipation of protracted caretaker periods.

A subsequent public lecture by the Solicitor-General (McGrath, 1999) arguesthat the cabinet manual is ‘descriptive’ rather than ‘prescriptive’ and lacks theforce of law. This accords with Wheare’s (1952: 179) view that distinguishes‘usage’ – ‘the usual practice’ – from ‘convention’ – ‘a binding rule’. Thisapproach, which is consistent with orthodox legal readings of responsiblegovernment, contradicts the views of most political scientists (for exampleTiernan, 2007) who assign considerable weight to the codification of ‘usages’into cabinet manuals. New Zealand’s leading political scientists express con-cerns regarding the lack of formalized details covering the behaviour of gov-ernments in the pre-election period. In this vein, for example, Boston et al.(1998: 10) have advocated that New Zealand study the Australian caretakerconventions, which commence with the dissolution of the House of Repres-entatives, rather than with the general election or the successful vote of noconfidence during the parliamentary term. In New Zealand, the period afterthe dissolution of the House, or in some cases the three months before the end of the parliamentary term, may see governments exercise restraintwith regard to ‘advertising’, or to ‘new appointments’. These broad usages are mentioned in the cabinet manual, but without the finer-grained detail of the coverage of the post-election period.

It is argued by some, however, that caretaker conventions ‘may’ applyafter the dissolution of the Parliament and until the election (see Boston et al., 1998: 10). The three-month period prior to the election is broadlygoverned by quite discretionary norms. This period is widely known as the‘interregnum’ amongst practitioners. An interview by the author with theformer Opposition Whip (5 May 2009) indicated that ‘interregnum’ was utilized by practitioners of the pre-MMP era. This covered a three-monthperiod before the election, and a period – normally short, after the election.Yet in New Zealand the practice remains of prime ministers announcing elec-tions well in advance of the dissolution of the Parliament and of conductingimportant parliamentary business after the election has been announced andwhen the country is in semi-election mode. In response, calls for the ‘policing’of the pre-election period are gathering momentum in New Zealand. Most discussion in the media relates to the post-election period (Creech, 1996;Cunliffe, 2005; English, 2008). For example, during the 2008 election cam-paign, and after the dissolution of the House, the Labour Finance Minister


announced a deposit guarantee scheme. Discussion ensued regarding the‘politicization’ of the Reserve Bank and it was stated that: ‘[Michael] Cullen[Labour Finance Minister] brushed off allegations … of politicization’ andthat in any case the opposition had in separate statements indicated supportfor such a scheme – given the economic climate (Kiwiblog, 2008). The PrimeMinister’s Office indicated it had placed ‘no obstacles in the way of the ReserveBank briefing the Opposition’. Others, however, commented that ‘caretakerconventions did not apply until after the election’ in New Zealand (Kiwiblog,2008).

Prime Minister Helen Clark’s September 2008 announcement of the generalelection date while the Parliament was in session had incensed the oppositionparties and generated debate over the interregnum period and its protocols.Until the Parliament was dissolved in October, there were extraordinary sceneson the TV news of MPs in full election mode while debating Parliamentarybusiness. A crucial Parliamentary committee was also sitting. Parliament wasdissolved two working weeks after the announcement. However, as tends tobe the case in New Zealand, the Parliament ran its full term.

Australia: Rules for restraining incumbency

While the development of rules governing the caretaker period after the dis-solution of Parliament and until the formation of a new government havebeen developed by the public service and generally been complied with, none-theless the capacity of governments to legislate for incumbency advantage in terms of the electoral law and other matters has generated controversy.Incumbency advantage has also played out in Australia in terms of the will-ingness of governments to call early elections; and more generally to ‘play’with the timing of announcements.

Hence, in stark contrast, to Helen Clark’s actions in New Zealand, around12 months earlier Australia’s Prime Minister John Howard announced anelection during the weekend break prior to a scheduled parliamentarysitting and the Governor-General immediately dissolved the Parliament.The election writs were issued several days later. Part of John Howard’s rea-soning was that it would be inappropriate and ‘wasteful’ of money andtime if the Parliament were to meet again. The term ‘recall Parliament’ wasutilized. He had behaved similarly in 1998, 2001 and 2004. The debate overelection announcements in Australia is complicated by the willingness ofpolitical leaders to call early elections, which does not necessarily tend tobe seen as illegitimate in Australia, especially when coupled in the publicmind with the Prime Minister’s capacity to initiate a double dissolutionwith the appropriate trigger.

However, John Howard’s actions in 2007 drew heavy criticism. Criticsmaintained that the shortness of time between the announcement and theelection would have negative implications for the quality of Australian

Marian Simms 101

democracy by preventing qualified people from voting (Costar, 2009). Otherissues raised regarding the Howard Government concerned the breach of care-taker norms by behaving as a ‘normal’ government after the announcementof the election date, and/or exploiting their executive authority for electoralpurposes. Such a concern was expressed by Tiernan (2008: 28) who notes thatin the weekend gap between Prime Minister Howard visiting the Governor-General on Friday 5 October 2001 and the Parliament’s dissolution on 8 October, the Prime Minister committed 1000 ground troops to support theUnited States’ air-strikes in Afghanistan. Howard was criticized subsequentlyfor not consulting the Opposition Leader Kim Beazley (Tiernan, 2008: 28).Jennings (1958: 158) states that such ‘strong leadership’ is entirely appropriatewithin a Westminster framework.

This section of the chapter will present data on Australian Prime Ministerialelection announcements since the mid-1980s and examine the question of the evolution of behavioural norms. It will argue that aspects of JohnHoward’s behaviour – where he ‘allegedly’ played fast and loose with caretakerconventions – can partly be interpreted as pay back for Bob Hawke and PaulKeating’s behaviour in the 1980s. It will also be argued that the special role of the Senate provides some legitimacy for a continuing political role for executive government.

Since the 1990s the Australian ‘norm’ has been for the Prime Minister to visit the Governor-General, call the election, and set in motion the process for the issuing of writs almost immediately. Recently the main complication in this streamlined process related to the changes to theCommonwealth Electoral Act (CEA) passed in 2006 which provided for theelectoral rolls to close three days after the issue of writs for updates, and forthe rolls to close for new enrollees at 8:00 PM after the writs were issued. In 2007 the Parliament was dissolved and the writs issued several days after the Prime Minister’s visit to the Governor-General’s residence inYarralumla (see Costar, 2009). Previously, under legislation introduced bythe ALP Government in 1983, the rolls closed seven days after the writs.

It is argued here that the Australian system already provides for a dif-ferent type of parliament than the Westminster norm by including a power-ful upper house. The Australian Senate was deliberately constructed to be a‘continuing’ institution and may have a right to meet after the prorogationof the Parliament, and its committees do meet after the dissolution of theHouse (Evans, 1997: 485). Its role has evolved especially since the introduc-tion of proportional representation for the 1949 election (van Onselen andErrington, 2005). The Senate’s evolving role has important implications forresponsible government which is predicated on the executive being ‘in andof’ the Parliament (Crisp, 1983: 34). Clearly once the Parliament has beendissolved then the executive is normally deemed to be in ‘caretaker mode’which continues through the election period and until it, or another min-istry, has the confidence of the Parliament. The Commonwealth’s Guidelines


(2007: 1) provide that the caretaker period ‘preceding an election for theHouse of Representatives’ commences with the dissolution of the House.The document is silent on the Senate’s role. Evans (1997: 483) notes thatfrom 1928–90 the proclamations dissolving the House also ‘discharged’Senators, despite lacking constitutional validity.

However in Australia the continuing nature of the Senate – except wherea double dissolution has occurred – means that the legislative function has a role independent of the executive. At times – such as on the eve ofthe 2004 election – Senate committees have used their work to embarrassthe government of the day, and to conduct electoral politics within theParliament. It is thus arguable that the existence of the Senate, coupledwith its party political nature, and the impact of proportional represent-ation, has engendered both a practice of the politicization of transition andan attempt to police that via the generation of guidelines and norms.

Another part of the post-1990s norm has been for the Prime Minister’sannouncement to be issued between parliamentary sessions, with signs of widespread acceptance of the idea that it would be wasteful for MPs toreturn to Parliament, or the idea that it would expedite campaigning forthem to stay out in their constituencies. As noted above, John Howardadhered to this process; but as in so many other matters the precedent hadalready been set by Labor. The 1996 election was called by Prime MinisterKeating almost two months after the Parliament had last met. He utilizedthe speedy timetable that John Howard also favoured.

As Weller (1985) notes, the actions of Australian Prime Ministers are con-strained by formal legislation. The Constitution provides for the broadtimetable for the dissolution of the Parliament, the issuing of writs andother aspects of electoral management (sections 43 and 44) and includesspecial mention of the role of the States in the issuing of writs for Senateelections (section 12). The Commonwealth Electoral Act provides otherconstraints, but has been considerably amended by recent Governments.The latitude available to Prime Ministers is still considerable. The exerciseof ‘informal’ power continues to be significant. Here the contrast betweenClark and Howard has been noted. Clearly Clark – known to have been a campaigning asset for the New Zealand Labour Party – was able to use the additional parliamentary exposure to her advantage, as well as needingto the additional parliamentary time to complete the legislative agenda,especially regarding the Emissions Trading legislation. Using the Parlia-ment as another forum for electioneering might be construed as anotherincumbency advantage.

In recent times the best comparator with Helen Clark would be Bob Hawke.In 1984 and 1987 he announced his intention to call an election well aheadof the formal processes. For example in his memoirs he refers to a ‘seven-weeklong, nightmarish’ campaign’: ‘On 8th October I announced a 1st Decemberpoll … The pressure for an election … was irresistible’ (Hawke, 1994: 274). The

Marian Simms 103

official campaign from the issue of the writs to the election was 35 days. The House met until 11 October and the Senate until 24 October. Much of the business was routine; although the on-going Costigan Royal Commissioninto the Federated Ships Painters’ and Dockers’ Union provided controversy,with the Opposition Leader, Andrew Peacock referring to Bob Hawke as a‘little crook’ because of the alleged links between the ALP, the unions andorganized crime that the Costigan Commission was uncovering. Hawke andhis advisors, or possibly Paul Keating and his advisors, had clearly learnt fromthe negativity of some of the 1984 parliamentary election campaign anddetermined to seize the initiative in the next election.

In 1987 Hawke again made an informal announcement on 27 May, ninedays ahead of the formal dissolution of the House. This time the Governmentseized the initiative in response to a question from a Labor MP – ‘Will theTreasurer and the Government be maintaining their commitment to fiscalresponsibility during the election campaign? Will the Treasurer continue toprovide specific figures and balance sheets and resist the temptation to followothers – I’m sorry to say, such as the Opposition Leader – in offering clichés.’The Treasurer, Paul Keating replied: ‘We will be going to the election cam-paign offering the Australian people a very precious commodity that is nowmissing in the political landscape and that is hope’ (Hansard, 1 June 1987).

The opposition replied in kind. John Howard, who had replaced AndrewPeacock as Liberal leader, asked a question without notice regarding theAustralian union involvement in blockading shipping heading towards Fiji.He asked: ‘Who is running the country, the elected government or the tradeunions of Australia?’ (Hansard, 1 June 1987). It is tempting to see Howard’sbehaviour in 2001 where he undertook important policy-making between theannouncement and the dissolution as ‘payback’ for Labor’s treatment of himin 1987: ‘… decision-making still rests with me because I am still PrimeMinister. You continue, of course, to govern and you certainly continue to take decisions within any existing established policy’ (quoted in Tiernan,2008: 28).

The strategy adopted going in to the next election was in accord with the current norm. Parliament was in summer recess when Hawke called the next election, and the Governor-General immediately dissolved theParliament and issued the writs. The shaky economic climate with highinterest rates and global economic uncertainty meant that the Labor gov-ernment was in a defensive mode, battling negative media coverage. Theonly plusses were continuing disarray in the opposition, where AndrewPeacock had toppled John Howard for the Liberal Party leadership and thesigns of growing support for the Greens and other minor parties whichcould return to Labor under the preferential voting system. According toHawke: ‘… the third party preference strategy was well advanced … (and)… polling underlined the environment as the second most importantconcern behind the economy’ (Hawke, 1994: 478).


Hawke presented as an embattled prime minister who did not wish tomeet the parliament, rather like a prime minister who knows that to debatethe opposition leader will provide her with opportunities. John Warhurst(2002: 12) reminds us that even unpopular opposition leaders do well indebates as was the case during the controversial 2001 election campaignwhen ‘Beazley won the debate … and the polls began to show Labor reduc-ing the Government’s big lead’. Yet previously in 1984 and 1987 Hawke’sconfidence carried him through both the parliamentary election campaignand into the Leader’s debate. David Butler made this comment on the 1984debate: ‘An honourable Bob Hawke felt bound by a promise made when he was unsuccessfully challenging Prime Minister Fraser to a televiseddebate in 1983 … Mr Hawke gave Mr Peacock ninety minutes of televisiontime before a huge captive audience, that saw him as being treated as equalwith the Prime Minister and demonstrating comparable authority andarticulacy’ (quoted in Hawke, 1994: 276–7). To use a soccer analogy, theLeaders’ debate was viewed as an ‘own goal’ by Hawke. This was a clear case of prime ministerial ‘overstretch’ where a risk taking style and anattempt to seek advantages from a position of incumbency strength failedto work.

Conclusion: Comparing caretaker conventions

Prime ministers have a range of tactics and opportunities but most involve adegree of unpredictability. This chapter contrasts two readings or variants ofpolitical leadership in the transition stages under Westminster: an ‘Australianvariant’ that focuses upon the need for prime ministerial ‘restraint’ after anelection announcement is made; and a ‘New Zealand variant’ which providesmore latitude, suggesting that a prime minister may take risks as part of herleadership style. Australian commentators want their prime ministers tobehave ethically once they are in election mode. New Zealand commentatorsare more relaxed and are prepared – in most instances – to leave this to prac-tical politics. The New Zealand focus on the post-election period predates the adoption of MMP in 1996 and emerged from a political crisis.

The issue here is whether using the Parliament to announce an electionis a good tactic, and/or bad politics. Certainly in Australia in 1984, the lengthof the campaign was seen as a negative feature by the ALP campaign director(personal communication with the ALP National Secretary). There is littleevidence regarding the electoral effects of the Labor Government’s election-eering in the Parliament. Turning to the ethical side, one concern might bethe government’s possible manipulation of political institutions for elec-toral advantage. It is this concern that has underpinned criticism of the gov-ernment’s use of public resources – such as postal and advertising budgets – for re-election purposes (see Young, 2004); and also underpinned criticismof government’s politicizing the public service during an election campaign,

Marian Simms 105

such as occurred in 2001 (see Tiernan, 2008). Yet the electioneering role ofthe Australian Parliament is more complex. While the Senate’s continuingand independent role is not in doubt, and would provide justification forits committees to play politics during campaigns, the fact remains that byconvention prime ministers are not senators. Yet according to some read-ings of Westminster (see Jennings, 1958) responsible government offersprime ministers wide latitude, and different times require different skillsand approaches. According to this view of governance, public opinion viaregular elections will make the appropriate judgements.

This chapter has argued that a key characteristic of the Westminster systemis the diffuseness of executive power, and that leaders who are intent onwinning will exploit such loose boundaries to their advantage. The transitionperiod, however it may be defined, permits wide latitude for leaders, especiallyin times of economic or political crisis; and that while changes to electoralsystems influence the way norms are expressed, changes to those norms aremore influenced by other factors, such as the constitutional framework andthe political culture.

The chapter shows that a mix of factors has combined to create dis-tinctive transition debates in Australia and New Zealand. These are: insti-tutional ones such as Australian Senate which has created the norm of acontinuous campaign and heightened the politicization of the pre-electionperiod, and the greater independence of the New Zealand Public Service;and cultural factors notably leadership styles where the emergence of leaderssuch as Bob Hawke (Australia) and Helen Clark (New Zealand) has led torisk taking including around transition matters. Further research as to whyrisk taking leaders emerge is a topic for another study.

Bibliography

Bagehot, W. (1867/1963) The English Constitution (London: Fontana).Bassett, M. and Bassett, J. (2006) Roderick Deane: His Life and Times (Auckland: Penguin

Books).Boston, J., Levine, S., McLeay, E., Roberts, N. S. and Schmidt, H. (1998) ‘Caretaker gov-

ernment and the evolution of caretaker conventions in New Zealand’, VictoriaUniversity of Wellington Law Review, 30. Accessed June 2009 http://www.austlii.edu.au/nz/journals/VUWLRev/1998/30.html.

Boys, M. H. (1997) ‘Speech by the RT Hon Sir Michael Hardie Boys to the 1997 HarknessHenry lecture: Continuity and change’, 31 July. Accessed July 2009 http://www.gg.govt.nz/node/471.

Clark, H. (2005) Foreword to the Cabinet Manual. Accessed December 2008 http://www.dpmc.govt.nz.

Commonwealth of Australia (2007) Guidance on Caretaker Conventions (Canberra:AGPS).

Costar, B. (2009) ‘New rules of the game’, Australian Cultural History. Special Issue:Kevin 07 – The 2007 Australian Election, edited by M. Simms, 27(2), 97–102.

Creech, W. (1996) ‘Tertiary funding allocations for 1997’. Media release, 6 November.


Crisp, L. F. (1983) Australian National Government, 5th edn (Melbourne: Longmans).Cunliffe, D. (2005) ‘Speech to the NZ Institute of Chartered Accountants’, Tax Matters

Archive. Accessed June 2008 http://www.taxpolicy.ird.govt.nz/news/archive.Davis, G., Ling, A., Scales, B. and Wilkins, R. (2001) ‘Rethinking caretaker conventions

for Australian governments’, Australian Journal of Public Administration, 60(3), 11–26.English, B. (2008) ‘Labour reminded about caretaker convention’, Press Release,

13 November.Evans, H. (1997) Odger’s Australian Senate Practice, 8th edn (Canberra: AGPS).Hawke, R. (1994) The Hawke Memoirs (Port Melbourne: William Heinemann).Jennings, I. (1958) The British Constitution (Cambridge: Cambridge University Press).Joseph, P. A. (2008) ‘MMP and the constitution: Future constitutional challenges’.

Paper presented at the symposium on MMP and the Constitution: 15 years Past, 15 years Forward. New Zealand Centre for Public Law, Victoria University ofWellington, 26–27 August.

Keating, M. (1999) ‘The public service: Independence, responsibility and responsive-ness’, Australian Journal of Public Administration, 58(1), 39–47.

Kiwiblog (2008) ‘Bollard admits risk in deposit guarantee scheme’. Accessed June2009 http://www.cc.bing.com/cache.

McGrath, J. (1999) ‘The Henry Harkness lecture: The Crown, the Parliament and theGovernment’, Waikato Law Review, 7, 1–21.

Malone, P. (2007) ‘Time to tighten caretaker conventions’, Democratic Audit of AustraliaDiscussion Paper, 3/07, Canberra: Australian National University.

Miller, R. (2005) Party Politics in New Zealand (Auckland: Oxford University Press).Rhodes, R. (2005) ‘Australia: The Westminster model as tradition’, in H. Patapan,

J. Wanna and P. Weller (eds) Westminster Legacies: Democracy and Responsible Govern-ment in Asia, Australasia and the Pacific (Sydney: University of New South Wales Press),pp. 129–52.

The Commonwealth’s Guidelines on Commonwealth of Australia (2007).Thomson, D. (1965) England in the Twentieth Century (Harmondsworth: Penguin). Tiernan, A. (2007) Power Without Responsibility: Ministerial Staffers in Australian Govern-

ments from Whitlam to Howard (Sydney: University of New South Wales). Tiernan, A. (2008) Caretaker Conventions in Australasia (Canberra: ANU E Press).van Onselen, P. and Errington, W. (2005) ‘Shock troops: The emerging role of senators

in House of Representative campaigns’, Australian Journal of Political Science, 40(3),357–71.

Warhurst, J. (2002) ‘International versus domestic issues’, in J. Warhurst and M. Simms (eds) 2001 The Centenary Election (St Lucia: University of QueenslandPress).

Weller, P. (1985) First Among Equals: Prime Ministers in Westminster Systems (Sydney:Allen & Unwin).

Weller, P. and Keating, M. (2000) ‘Cabinet government: An institution under pressure’,in M. Keating, J. Wanna and P. Weller (eds) Institutions on the Edge: Capacity forGovernance (Sydney: Allen & Unwin), pp. 45–73.

Western Australia, Premier and Cabinet (2008) Caretaker Conventions (Perth: Govern-ment Printer).

Wheare, K. C. (1952) Modern Constitutions (Oxford: Oxford University Press). Young, S. (2004) The Persuaders: Inside the Hidden World of Political Advertising (Sydney:

Pluto Press).

Marian Simms 107

Part II

Understanding Successions

7Seeking and Keeping the Hot Seat: A Comparative Analysis of PartyLeader Successions Matthew Laing and Paul ’t Hart

Into the hot seat

Seen from the outside, the leadership position in a political party or gov-ernment is coveted by many. Yet party leadership and, possibly in its wake,a prime-ministership or a presidency are ‘hot seats’ in more than one sense.They are not just hot in terms of the potential for power and authority theybring to those that occupy them; they are also hot in terms of attractingcompetition and controversy. Leadership of a party is more often than nota precarious possession: not only do political leaders lead their lives con-stantly in the public eye with all the drawbacks that this entails, but thereare also plenty of people ready to criticize their performance. There is noshortage of potential competitors either, all whom are brooding on howand when to take over the top job.

In dealing with leadership issues, political parties generally will want to balance stability at the top with the leverage to rid themselves of unpopular, ill-respected or otherwise unsuitable leaders. They also need todecide how inclusive they want to be when it comes to making decisionsabout the leadership. In this light they need to consider the pros and cons of different modes for selecting (Marsh, 1993; Davis, 1998; LeDuc,2001) and removing (Weller, 1983, 1994; Alderman and Smith, 1990;Quinn, 2005) leaders. This means choosing between different types of‘selectorates’ (actors and bodies involved in selecting and dismissingleaders), varying from the closed circle of elite consultation to party roomvoting to leadership conferences to all-member ballots (see also Heppell,2008, 2010).

As noted in the introductory chapter, much of the prior research ondemocratic leadership succession focuses on the role of these institutional

111

mechanisms. This line of inquiry is pursued further in Cross and Blais (thisvolume). In our contribution, we turn to the other side of the coin: the roleof key protagonists in succession episodes – incumbent and aspiring partyleaders – and the ‘temperature’ of the resultant process in which leadershipissues are being settled. Taking into account the party’s rules and traditions,the political context at the time, and their relative standing within andoutside the party, leadership hopefuls face delicate choices in manoeuvringto overcome the obstacles that confront them in what is more often thannot a long and arduous road to the top. Moreover, once in office, partyleaders encounter a new set of choices, evolving around the balancebetween two contradictory desiderata: preserving party unity by projectingcontinuity versus asserting authority by highlighting the differencesbetween the outgoing and the new regime.

We explore the extent to which the actors in question can and do shapetheir own fates despite these institutional and contextual constraints, and the degree to which new leaders can become the victims of the veryadversarial nature of the process through which they have gained the top job. To look at this question, we have used a dataset of leadership successions in combination with cases studies and in-depth research. Interms of scope, the dataset attempts to capture as many leaders of partiesthat held office at least once between 1945 and 2009 in countries definedas democracies in Arend Lijphart’s Patterns of Democracy (1999). Not allcould be included, either because of a lack of available information for anon-native speaker to collect (e.g. Mauritius), the institutional impossibilityof establishing a single political party leader in some countries (e.g. Italyprior to 1992), or a change in their democratic status since the publicationof the book (e.g. Venezuela). Ultimately, 23 out of Lijphart’s 36 wereincluded.1

The database comprises 515 cases of succession since 1945, covering 66political parties in 23 countries. We coded each case on 36 different vari-ables, covering the main protagonists’ demographic characteristics; theduration as well as key characteristics of the beginning and end of theirleadership tenures; party and party system variables; and outcome vari-ables.2 Only a limited part of the dataset will be used here and we will limitourselves to basic descriptive and correlational analyses. In addition wedrew on high profile cases well-documented in the succession literature andthe popular press for illustrative and explorative purposes.

Assessing succession outcomes

We first turn to the dependent variable: succession outcomes. These weremeasured using two key indicators: the longevity of the leader and theparty’s first post-succession election results under a new leader in terms ofthe change in seat share and vote share. In our sample, leadership longevity

112 Seeking and Keeping the Hot Seat

ranges from a little as 47 days in the case of Sir Donald Sangster’s leader-ship of the Jamaican Labour Party to an epic 12,988 days (>35 years) in thecase of Dom Mintoff’s leadership of the Maltese Labour Party. Although thearithmetic mean would postulate that leaders on average lead their partiesfor five to seven years, the highly skewed and kurtotic nature of the long-evity distribution means the large majority of leaders either far exceed orfall far short of that time period (see Figure 7.1). Nearly a quarter of leaders(100 of 452) did not make it to the end of a second year in office and about35 per cent (162 of 452) did not reach beyond a third. A minority of coun-tries such as Australia and New Zealand has had three-year electoral cyclesfor all or most of the research period. Yet this can hardly explain such startling results. At the other end, an equally remarkable near 30 per centof party leaders survived for ten or more years in office (125 of 452). Whatwe see seems to be a crash or crash-through pattern of party leadership:most new party leaders either leave office quickly or complete a relativelylong term.

This pattern of distribution is relatively uniform when controlling fornational political cultures, electoral systems and party types. In terms ofnations, only Malta, Trinidad and Tobago, Jamaica and Botswana (which con-tribute only a very small number of cases between them) prove to deviate

Matthew Laing and Paul ’t Hart 113

60

40

20

00.00 10.00 20.00 30.00 40.00

Histogram

Fre

quen

cy

duration (years)

Mean = 7.22 Std. Dev. = 6.585

N = 452

Figure 7.1 Party leader longevity distribution

from the above distribution pattern significantly, and in all four cases theleaders of parties in these countries stay in office much longer than average.Also noteworthy is Japan – though it does not deviate as much as the otherfour cases, it is a negative outlier, with many of its leaders holding onto officefor shorter periods than counterparts in other countries.

In terms of electoral systems there was some expectation that the leader-centric and rough-and-tumble nature of the Westminster system mightdepress leader longevity. Not so in our dataset. Leaders from Westminstersystems were somewhat more likely to exit within two years (24% v. 19%)but served comparable mean terms (7.3 years for Westminster vs. 6.9 yearsfor non-Westminster). Similarly, there are no significant differences in pat-terns of leadership longevity across two-party, multi-party, or so-called‘two-and-a-half-party’ systems – all three exhibited a distribution patternvery similar to the general pattern outlined above, with no system muchmore likely to produce long- or short-term leaders. There is some deviationin the case of simple two-party systems, with more medium-term leaders(and hence a less skewed and kurtotic distribution) than the other systems,however overall the difference is relatively small. The major outlier is theuncommon party arrangement (representing less than 5 per cent of the cases)– the Dominant Party system. Countries with a single dominant party oper-ating either feature leaders much longer or rather shorter than the norm – Japan, Luxemburg and Botswana being obvious examples.

We measured the electoral impact of succession in terms of the percentagechange in the total vote share and seat share achieved by a party at the firstelection after the selection of a new leader when compared with the previ-ous election. Post-succession election results were nearly evenly dividedbetween negative and positive outcomes in a fairly normal (thoughstrongly peaked) distribution, with the average change in vote share veryslightly positive at +0.0629 and the average change in seat share alsoslightly positive at +0.4162. There is little evidence in these figures or in thegeneral distribution of the results to suggest then that a change in leader is likely to deliver a good or bad outcome – that is to say it is unlikely achange of leader in and of itself is a necessarily positive or negative factorat subsequent elections. The control variables do not have much of an impacton post-succession election results. In some countries (such as Canada andBarbados) the change in seat share can be more dramatic due to the designof the electoral system, but the more reliable figure of vote share does notchange very significantly across systems.

The two measures of succession outcomes may be correlated. In a worldof rational political actors, new leaders are generally expected to help improvetheir party’s showing in the polls and at the next election. This thenbecomes a requirement for them to be seen as successful and be able toretain office (see Andrews and Jackman, 2008). There is in fact a weak butstatistically significant correlation in this dataset linking the difference


between the incumbent’s last election result and the successor’s first on theone hand and the successor’s tenure length on the other (for vote share difference, Pearson’s R = +0.213, sig. p < 0.01; for seat share difference,Pearson’s R = +0.243, sig. p < 0.01). However, the numbers do not speak for themselves, as contemporaries interpret the election outcomes in view of the prevailing prognoses, circumstances and climate, adjusting theiraspiration levels accordingly. Some parties may improve their number ofseats considerably from one election to the next, yet still their leaders will be regarded as not very successful, because the aspiration level was for them to do even better. But the reverse also holds true: parties may lose seats under the new leadership’s first election, but that fact may notnecessarily be held against the leader if contextual factors are judged to have conspired against any better result from the outset. In coalitionsystems moreover, parties can ‘lose’ the election but be perceived to have ‘won’ the subsequent coalition formation process through deft post-electoral manoeuvring.

We should place these patterns of succession outcomes in a somewhatbroader perspective. The nature of the political process has been widelydepicted as having undergone significant changes during the post-World War II period. Key megatrends include secularization, dissolution or recon-figuration of social cleavages, a decline of partisan loyalty among voters,decline of trust in political institutions, and the rise of televised and web-based political competition. These have combined to produce what has been called a growing ‘personalization’ of the political process, and certainlyof voter choices (Blondel and Thibault, 2009; Aarts et al., 2010; cf. King,2002). People do appear to vote less and less for parties per se, and more andmore for (or against) individual leaders than they used to. We take this toimply that the stakes for leadership succession have risen over time. The elec-toral fate of the party depends increasingly on ‘picking a winner’ in for theleadership, and consequently the leader’s fate rests more on the ability todeliver electoral results and maintain public popularity than in previousperiods. In an age of constant opinion polling and volatile public sentiment,this means too that a leader who fails to deliver can expect punishment fasterand more severely.

Indeed, in our dataset there is a significant negative correlation (Pearson’sR = –0.457, sig. p < 0.01) between the historical era when the leader tookoffice and their longevity. Contemporary leaders serve for far shorter thantheir post-war counterparts. Comparing two similar time periods (1945–65vs. 1989–2009) at opposite ends of our research period, interestingly thecrash/crash-through distribution pattern still (roughly) holds, though mid-points and range of the distributions are radically different. Leaders in thecontemporary group served a median time in office of 1037 days, almosttwo-thirds less that of the post-war group (2973 days). Just fewer than 25 per cent of leaders in the post-war group served in excess of 6000 days


– no leader in the contemporary group accomplished this, and just 6 percent led in excess of 4500 days. So if anything, the stakes for making the‘right choices’ have increased in recent decades. Let us now turn to theroles and impact of key protagonists in succession episodes (aspirants andincumbents) and the resulting climate of the succession process that resultsfrom their postures and choices.

Explaining succession outcomes: Protagonists and processes

Aspirants and new leaders

Prior to a succession episode, members of the party elite who have eitherpublicly committed to having leadership ambitions or are touted to enter-tain such ambitions, face three strategic choices when a succession triggermaterializes. The first is whether or not to run at the present time. This isnot a vacuous choice. Although succession triggers – bad polls, backbenchrevolts, factional wars, ill health or death of the incumbent – do not comealong all that often in most parties, even the most ambitious aspirant real-izes that some occasions are better than others for them to try and take thehelm of the party. Certain triggers may come too early or for that mattertoo late in terms of their political capital with the power brokers and therank and file of the party. Moreover, some points in the political businesscycle are far less attractive than others to be cast into the leadership. Forexample, long-term governing parties that have recently lost govern-ment are hardly a promising prospect, or, in proportional representationelectoral systems, parties that are consistently outmanoeuvred in coalitionformation processes. Likewise, parties that suffer from bitter internal divisionson one or several key long-term political issues (for example, the BritishConservatives on ‘Europe’ during the 1990s and beyond, as well as variousEuropean social democratic parties on nuclear rearmament and nuclearpower generation in the 1980s and 1990s) may present poisoned chalicesto leadership aspirants. In all of these circumstances, key leadership hope-fuls may decide to forego running at the time when a succession triggerpresents itself.

For those that do decide to run at a particular time, two further choicesare pivotal: how to position themselves towards the incumbent leader andtowards the party’s current platform. All options have potential benefitsand risks. Take for example the relationship to the incumbent. Aspirantsmay seek to become his preferred replacement in a mutually agreed uponstrategy of ‘managed transition’. But they may also deliberately presentthemselves as an overt critic of the existing leader, and promise ‘new depar-tures’ and ‘different leadership’ to constituencies. In the former case, theymake their political fate dependent upon the state of mind and loyalty ofsomeone who is not necessarily motivated to act nobly and altruistically.When they choose a competitive mode, candidate leaders expose them-


selves to the ire of the incumbent and his entourage – powerful enemies tomake. Moreover, dependent upon their tone and timing they run the riskof splitting the entire party and hurting its political standing, at least in theshort run. This will leave them with even more enemies within the partyestablishment.

Aspiring leaders who want to take the confrontational route not onlyneed to believe strongly in the popular appeal of their candidacy but alsoneed a coincidental succession context that makes this political gambleseem potentially viable, perhaps even necessary. This occurs when there iswidespread or at least growing public (and/or media) dissatisfaction withthe existing leadership, which would be most pronounced in the case of asuccession episode triggered by a scandal involving the incumbent or anincumbent having lost an election widely regarded a winnable.

In addition, aspirants also face a key tactical choice when a successiontrigger materializes (particularly in case of an exogenous trigger): maketheir ambitions known, and risk mobilizing their own opposition; or besilent about their plans at the risk of handing the initiative to others thatdo ‘come out’. In this sense then, an aspirant can run for a leadership posi-tion without waging an actual campaign. Incumbent leaders know this alltoo well: the pledges of continued support given to them by even theirclosest colleagues are always conditional and provisional, and sometimesclearly hypocritical. This can induce paranoia in them at times when theirpolitical stocks are low.

Moving to the post-succession stage, newly elected party leaders mustdecide what to do with their new role. Some come to the job meticulouslyprepared; others are cast into the role quite unexpectedly. Some are policywonks with detailed ideas about the future of the party programme and itselectoral platform in the next election. Others, having concentrated theirefforts on getting enough votes within the party organization to get in,now primarily begin to focus on becoming known and liked enough by therest of the electorate to win the next election. Whichever inclination theymight have, none of them can escape the key question of how to relate to,and sometimes deal with, their predecessor.

The new leader who ran as a ‘loyalist’ type aspirant has the problem ofcoming to be seen as ‘his own man’ without losing the support of theformer leader and his coalition. Shortly after taking over from MargaretThatcher, John Major had to endure opposition jibes that he was the ‘nochange, no-majority prime minister’, and touting him ‘to stamp his ownauthority on the government he has inherited, to provide that he is not“son of Thatcher”’ (Seldon, 1998: 132). Major took that advice and startedprobing new departures in policy and political style, only to earn himselfthe animosity of his predecessor who would undercut him at every possibleopportunity and weaken his position within the party. By comparison, suc-cessors who take control after both the old policies and the old leader have


been firmly discredited have a much easier job. The necessity for renewal isuncontested, and putting distance between the past and the present is notso much a delicate option but an elementary political necessity.

The toughest challenge is perhaps faced by new leaders who ran as reformersor revolutionaries. Established political parties tend to be creatures of history.They tend to cherish their past, their traditions, and their historical program-matic cores. Long-standing members of a party’s ruling oligarchies tend toinvoke these to protect their turf, patronage networks and key policy com-mitments. To get them to change course and implement reform is a dauntingtask for any leader (Panebianco, 1988; Katz and Mair, 1995). But it is evenmore daunting for a leader who has recently been voted in, and still has toestablish himself within the party, its closest social and political allies, andamong the general electorate.


Table 7.1 Strategic choice for leadership aspirants and new leaders

Attitude towards Attitude towards incumbent

party platform Supportive Critical

Supportive Loyalist Rival(Louis St. Laurent, (Jenny Shipley, New Zealand Canadian Liberal Party, 1948) National Party, 1997)

Critical Reformer Revolutionary(Tony Blair, UK Labour Party, (Gough Whitlam, Australian 1994) Labor Party, 1966–67)

Table 7.1 summarizes the main strategic options aspirants and new leadersface. Assuming that parties generally are path-dependent, risk-averse organ-izations, we can therefore put forward the proposition that: (1) Loyalists are the most likely, and revolutionaries the least likely, to experience a success-ful succession (except for periods immediately following major election lossesand/or losses of government for the parties when reformers are more likely thanloyalists).

For each leader in the dataset we sought to code them corresponding to the typology. We coded conservatively: an aspiring/new party leader wasonly classified as ‘critical’ in terms of platform and predecessor only whenthis was unambiguously the case. Those who took a mixed or moderatestance were generally coded as supportive on those counts. We did notcode cases where there was insufficient information to make a judgement.Table 7.2 shows that half of our leaders could be considered to be loyalists,with the other three categories sharing the remaining cases in roughly equalproportion, though leaders considered to be purely rivals are the rarest of the types.

Table 7.3 describes the distribution of succession outcomes (longevity; electoral results in first election).3 It is immediately clear that ‘rivals’, with atrimmed mean of serving time of just 3.72 years and 50 per cent serving lessthan two years in office, are clearly least likely to secure a smooth successionfor themselves. The distribution of longevities for the rival category is heavilyweighted towards an early exit. When they do make it to the next election,they have not been overly successful but neither have they been disastrous.


Table 7.2 Distribution of aspirant/new leader types

Frequency Percent Valid Percent Cumulative Percent

ValidLoyalist 238 46.1 54.3 54.6Reformer 85 16.5 19.4 74.0Revolutionary 64 12.4 14.6 88.6Rival 50 9.7 11.4 100.0Total 437 84.7 100.0

Missing 79 15.3Total 516 100.0

Table 7.3 Succession outcomes for successor types

Post-Succession Trimmed Outcome Mean Mean Median Variance Skewness Kurtosis

Vote Share Change –1.0528 –0.9373 –0.4350 39.973 –.422 2.177Seat Share Change –1.8565 –1.0452 –0.5700 119.616 –1.829 7.935Tenure (Years) 5.4910 4.9569 3.7288 25.118 1.686 3.284

Vote Share Change 0.8316 0.7923 1.2000 25.245 0.066 0.476Seat Share Change 2.5291 1.7521 0.6000 100.624 1.796 6.642Tenure (Years) 9.5978 9.2339 8.8000 37.728 0.751 0.203

Vote Share Change 2.8214 2.5024 2.7000 34.084 0.834 0.697Seat Share Change 4.7376 3.9571 1.0000 120.367 1.383 2.062Tenure (Years) 10.5991 10.0025 9.8301 57.221 1.153 1.738

Vote Share Change –.1225 –.3618 –0.8000 38.516 0.800 2.431Seat Share Change 0.2692 –.3169 –0.6150 95.306 0.813 0.986Tenure (Years) 3.4844 3.0782 1.9986 12.192 1.799 2.848

Lo

yal

ist

Typ

eR

efo

rmer

Rev

olu

tio

nar

yR

ival

Conversely ‘reformers’ experience the most successful successions. The reformergroup has the closest thing yet found in any our testing to a standard dis-tribution, with the most significant clustering of leaders occurring close to the mean. Only 17 per cent of leaders in this category do not make it to thefourth year (again, much lower than for any other grouping studied in thischapter). Although their mean tenure length and electoral results are not ashigh as for the revolutionary group, the sheer consistency of relatively strongresults in this group make it the most successful.

‘Revolutionaries’ are also relatively successful, with the highest mean serv-ing times and electoral results post-succession. However the distribution oflongevities is closer to the crash/crash-through model observed earlier and thestrong results of some are offset by really poor results of others. Finally, loyal-ists are a slightly underperforming group, though as expected for the defaultgroup, their succession results and respective distributions are very similar tothe population as a whole.

Revolutionaries and reformers are therefore the most successful groupswhen compared against the dataset as a whole (and rivals the most unsuc-cessful). What is more, the pattern of consistency in their success suggeststhat it is not dependent on specific circumstances, though some contextfactors do make a difference. For example, revolutionaries and reformersstand little chance of succeeding in governing parties: of the 98 successionswhich were both for party leader and head of government, and there wassufficient data to code for successor typology, just 18 (18.4%) producedeither of those two types. Rewarding loyalty seems to be the norm for in-government successions: in 71.4 per cent of such successions loyaliststook over, and in 10.2 per cent rivals (who support the party platform thoughnot the incumbent). In contrast, opposition parties were more than twiceas likely (41.5%) than governing parties to take a chance on reformers orrevolutionaries.

Clearly, parties in electoral trouble are more willing to change in order to appeal to the electorate once more, and when they do secure the leader-ship reformers and revolutionaries are often able to lift their parties fromperiods of lacklustre party performance. Conversely, although they mostoften manage to secure succession, contrary to Proposition 1 loyalists donot seem to have better chances of succeeding as party leader. Even thoughthey are the overwhelming favourites to succeed (particularly in governingparties), our data suggest that in-government successions are mostly unsuc-cessful and even in non-governing parties loyalist new leaders don’t do well.What works to get into leadership is different from what it takes to succeedonce in the hot seat.

Although Proposition 1 is not well supported, a few interesting patterns haveemerged from our analysis of it. Firstly, the difference in success between rivalson the one hand, and reformers or revolutionaries on the other, provides aninteresting observation. Conflict between predecessor and successor seems a


significant factor only when the successor is not also trying to affect majorchange to the party as a whole. The strong electoral results banked by bothreformers and revolutionaries indicates that the public are much more inter-ested in platform change than party intrigue, and strong electoral results are usually sufficient to quell party critics and settle the matter of a disputesuccession. Australian Labor Party leader Gough Whitlam was the classicrevolutionary – fiercely committed to party and platform change and anongoing thorn in the side the Labor old guard, his assumption of the leader-ship in 1967 heralded a significant and internally contentious change in theparty. Yet despite his initially rocky years as leader, the ALP’s strong results atthe 1969 election quieted his critics and consolidated his leadership.

Unlike reformers, revolutionaries have less room for error in their initialleadership and perhaps their main drawback is increased animosity andeventual ‘payback’ from within their own parties (discussed further below).By avoiding interpersonal acrimony, reformers on the other hand may havegreater room to move and be given more time to allow their reforms to takeeffect. For example, Guido Westerwelle’s ascent to the leadership of the FreeDemocratic Party of Germany (FDP) in 2001 marked a period of significantreform for the party. However the party was slow to reap the benefits ofchange, garnering only slightly improved results at the 2002 elections. YetWesterwelle was internally well positioned within the FDP and maintainedthe support of his colleagues and the more traditional wing of the party longenough to see unprecedented back-to-back electoral gains in 2005 and 2009,followed by a return into government. The major risk for reformers (or forthat matter, revolutionaries) is to move their parties in the ‘wrong’ direction,e.g. Michael Foot’s retroactive socialist reform agenda for the UK Labour Partyproving to be a disaster at the 1983 elections. Although many parties willinstinctively chose loyalists, particularly in high stakes situations (such aswhen the party is in government), it is not clear that this choice is manifestlythe safer one than more reformist alternatives.

Incumbents

Unless succession episodes are self-initiated, or imposed on them by forcemajeure or non-negotiable term limits, incumbents need to make up theirmind whether to stay or go. If they decide to go, they need to considerwhether and how to influence the selection of a successor. If they decide tostay, they face the tactical choice of how to best achieve that goal: byavoiding or delaying any formal decision moment whilst working hard toreassert their authority within the party, or by calling their detractors’ bluffby pre-emptively forcing a decision sooner rather than later? We shouldkeep in mind in all of this that the degree to which incumbents’ choices orpreferences impact upon the course and outcomes of succession processesis largely contingent upon their standing within the party and among thepublic at the relevant point in time.


In short: the stronger the incumbent’s authority and the stronger hisdetermination to stay, the lower the likelihood that succession will occur.At the same time however, when high-authority incumbents decide to stepdown of their own accord, they – unlike low-authority incumbents – maystill be able to exercise considerable influence over the party’s selection of the successor. Table 7.4 provides a typology of incumbent’s strategic pos-itions vis-à-vis a looming succession challenge. From it we can derive the following proposition concerning the viability of the various strategicoptions incumbents face at such times: (2) Ceteris paribus, ‘winners’ are theleast likely and ‘losers’ the most likely to be forced or pressured to resign by theirparties.


Table 7.4 A typology of incumbent leaders’ positions vis-à-vis successionchallenges

Hold Over Level of External Support

Own Party Strong Weak

Firm Winner OligarchImpossible to challenge Will put up a hard fight until (Tage Erlander, Swedish Social electoral realities are really Democratic Party, 1946–69) overwhelming

(Arthur Calwell, Australian LaborParty, 1960–67)

Weak Maverick LoserVulnerable to changing public Ripe for the killsentiment (Michael Foot, British Labour (Willy Brandt, German Social Party, 1980–83)Democratic Party, 1964–74)

In testing the proposition, we did not code for the degree of controlleaders have over their parties, and measuring their relative externalsupport (outside of elections) is currently beyond the scope of our currentdataset. The true nature the leader’s standing within their own party oftendoes not become apparent to outsiders until a major event occurs whichcan serve as a litmus test one way or the other (e.g. leadership challenges orparty conferences). Secondly, for cases occurring before the age of constantpublic polling, determining leaders’ public popularity produced a highnumber of missing cases. So the account that follows will be qualitativeand exploratory.

The 163 total cases in the dataset where the leader has been either forcedto leave or has done so under duress are quite diverse. However, it is rareindeed to find a case where a leader was pressured or forced to resign whilstsimultaneously popular with the public and well positioned within their

party. When examples are found, they tend to be unusual or unexpectedcases. For example, the 1967 resignation of Liberal People’s Party (LPP)leader Dr Bertil Ohlin came at a time when Ohlin was still a popular elderstatesman in Sweden and commanded a great deal of respect and admira-tion within his own party. Ohlin had led the LPP since 1944 and hadcarried the party to record election results. Although results in the latterhalf of his leadership were not of the calibre of the first half, even the worstresult under his leadership (in 1964) was much better than any result receivedprior his leadership (and as time would tell, after it as well). An intellectualgiant and future Nobel Prize winner in economics, Ohlin also occupied thepinnacle of policy competence in his party and commanded the support ofthe rank and file. In the late 1960s a movement had emerged in the youthwing of the party. Future LPP leaders Ola Ullsten and Per Ahlmark began topressure for change and renewal within the party and sought to draw Ohlininto retirement. Although Ohlin still commanded strong support fromwithin the parliamentary party and with the Swedish public, he was 68 anda series of political successes by the youth wing despite his opposition wasenough to trigger his resignation in 1967. However, it must be said that thethreshold for Ohlin’s resignation was in this case much lower than usual – it is unlikely a direct challenge to Ohlin’s leadership from the youthmovement would have succeeded.

As predicted, ‘losers’ are well represented within the duress/forced exitsubset. Leaders who come to office with little power within the party and whofail to gel with the public often do not last long and are forced out. Althoughan honest and hard working politician, Enneüs Heerma’s leadership of theDutch Christian Democratic Appeal was doomed almost from the start. Along-time, mainly ‘technocratic’ cabinet minister, he came into office in thewake of the party’s dramatic 1994 election defeat after which it was relegatedto opposition for the first time in history. The party room was dispirited andfractious. Heerma’s style and message did not resonate with the Dutch publicas he struggled to make headway in the polls. The party swiftly made his pos-ition untenable and he was compelled to resign before fighting a single elec-tion as the party turned to the up-and-coming Jaap de Hoop Scheffer (soon to meet a similarly ignominious end). Leaders like Luis Marques Mendes(Portugal-Social Democratic Party), Wolfgang Schauble (Germany-ChristianDemocratic Union), Ian Duncan-Smith (UK-Conservative), Binyamin Ben-Eliezer (Israel-Labour) and Simon Crean (Australia-Labor) are similar examples.They were leaders who came into their parties with moderate or weak stand-ings within their own party and could not make inroads into the party’smomentarily weak electoral standing. They all found themselves caught up ina negative spiral, and all were ousted pre-emptively. Political events quicklyovertake these leaders and few will willingly chose to cut their own leader-ships short before they have had a chance to prove themselves (a chance notalways awarded by their parties).


Once politically more efficacious leaders whose internal and/or externalpolitical capital is shrinking have more opportunity to resign before pres-sure mounts. A few like Canadian Liberal leader Pierre Trudeau will see thewriting on the wall and choose to retire whilst still in reasonable politicalshape. Some like Barbados Labour Party leader and Prime Minister OwenArthur wait until the evidence is clear (like a poor election result) beforeresigning voluntarily rather than face a slow decline and wait for the partyto turn against them. Others will simply refuse to accept the new realitiesand more or less compel their parties to move decisively against them, as inthe case of Canadian Progressive Conservative leader John Diefenbaker and,to some extent, Margaret Thatcher. With a political ‘loser’ in the leadershipposition, party and political pressure for replacement is very quick to follow.

An interesting question is which of the leaders with a strong grip on onlyone of the two elements in our incumbency matrix is more likely to avoidquick and forced exits. The coded cases in the dataset indicate that formajor parties, a lack of public popularity can rarely be compensated for bystrong control over the party. ‘Oligarchs’ are more common and have abetter survival rate in smaller parties, owing perhaps to the greater ease atwhich a leader can maintain personalistic control of party platform andapparatus. This is particularly so in parties at the far ends of the spectrumwhere ideological purity is deemed more important than electoral marketshare or government participation. For example, Alvaro Cunhal (Portugal-Communist Party) and Charilaos Florakis (Greece-KKE) maintained controlover their far-left parties for decades seemingly unimpeded by a completelack of electoral popularity.

Conversely, in electorally successful but internally weak or divided majorparties leaders appear to be more capable of survival, though only as longas they remain popular with the voting public. ‘Mavericks’ like WillyBrandt (Germany-SPD), Sonia Gandhi (India-Congress Party) and JunichiroKoizumi (Japan-LDP) challenged their parties’ traditional tenets and estab-lished elites. They made enemies within their parties and faced regularinternal debates about their continued leadership. Yet they successfullydeployed popular support as an aegis against internal enemies and poten-tial challenges. In smaller, ideologically zealous anti-establishment parties,leaders face conformity pressures to be clearly seen to be ‘one of us’ (socialpsychologists refer to this as ‘relative prototypicality’, see Turner, Subasic,and Reynolds, 2008; Subasic and Reynolds, this volume) and are less ableto use broader popularity as a means of increasing their idiosyncrasy credit.

As highlighted earlier, with our current dataset and coding we can onlystart to examine this particular proposition and some of its implications.We do find that the proposition is well supported by available cases, butoverall the more interesting questions related to our matrix for incumbentslies with the more hybrid cases, and as such present an interesting area forfurther research.


Succession processes

Here we want to investigate whether the dynamics of the succession processgenerated by the postures and behaviours of the main protagonists and theirsupporters themselves may have an influence on the fate of whoever ends upin the leadership position (it should be remembered that succession episodesmay produce ‘non-successions’ as well, in which incumbents stare down theirchallengers or otherwise get reappointed).4 It seems plausible to assume thatintensely competitive, secretive and vitriolic leadership campaigns leave deepwounds among the protagonists and their supporters and hardly send reassur-ing messages to members and voters alike. They are thus are more likely toresult in continued leadership instability. In contrast, mutually respectful andmore inclusive succession processes are more likely to provide firm and broadmandates for the resultant leader (cf. Cross and Blais, 2008; Denham andO’Hara, 2008; Kenig, 2009). Hence our third proposition: (3) The more conflict-ridden the succession process, the lower the likelihood of successful succession.

An exemplary case that fits the bill entirely can be found in the DanishConservative People’s Party (CPP). After the surprise resignation of CPPleader and prime minister Poul Schlüter in 1993, a power vacuum openedup within the party. Long-time rivals Justice Minister Hans Engell andFinance Minister Henning Dyremose staked their claim to the party leader-ship. Dyremose won, but the conflict between the opposing wings of theparty continued to plague him. Within nine months Dyremose was toppledby Engell in a party room coup, after which he quit politics. The factional dis-putes within the party persisted however – Engell would maintain control forthree years before he was himself compelled to resign after a drunk-drivingincident in 1997. The race to choose his successor reignited the severe internaldisputes of 1993, with Per Stig Møller narrowly triumphing. After five years ofvery public infighting the voters had lost confidence in the party, and at the1998 Danish elections the CPP received their worst election result in 25 years.Internal disputes flared up once more in the wake of the elections, and fac-tional opponent Pia Christmas-Møller replaced Per Stig Møller in 1998. Herown contested leadership lasted just over 18 months before a final contestedleadership battle delivered Bendt Bendtsen to the top post. Bendtsen wasfinally successful in unifying the party and ending the factional politics whichhad destroyed the party’s public image and seen five leaders in six years.

How typical is this case? As a proxy to coding the climate of the entire succession process, we assessed the degree of conflict in the chief triggers ofsuccession. If the case was unclear or there remained argument as to what thetrue motivations for the succession were, it was left un-coded. Furthermore, incases where it was difficult to establish a single trigger, a secondary trigger wasadded to try and capture that. The result is 452 cases where a trigger could besatisfactorily determined, using a 16-category coding grid, dichotomized forour present purposes. Under high-conflict triggers came: inter-personal rivalry,internal party discontent or factional intrigue, internal policy disputes, and


inter-coalition conflict. The other triggers (ill health, election losses, and soon) were classified as low conflict. Using this breakdown, conflict was theprimary cause in 65 out 452 cases, and a secondary cause in 57 (see Table 7.5).

Table 7.5 suggests that the correlation between high-conflict succession trig-gers and succession outcomes is weak. Descriptively, in the 89 valid cases of aconflict succession, 27 per cent of successors (24) lost office within two years,and on average served six years as leader. This compares with about 19 percent of non-conflict successors (70 of 363) and an average of 7.5 years. Thecorrelation between conflict and leader longevity is negative as predicted butsmall (Pearson’s R = –0.088, sig. p < 0.05). Looking at the differences in votingshare under the next election, the 59 high-conflict successors who survived to the next election could expect on average a small but negative swingagainst them – about –0.34%, compared with +0.17% for low-conflict suc-cessors. Consistent with the correlation above, the distribution patterns areonly slightly different – 58 per cent (34 of 59) of conflict successors sufferednegative swings at their next election, compared with 49 per cent non-conflictsuccessor (109 of 220). Consistent with the longevity finding, a slightly higherproportion of high-conflict successors lost office before facing an electionthan non-conflict successions (34% to 25%). Neither of these correlations is statistically significant, however. We then revisited incumbent exit modes


Table 7.5 Primary and secondary succession triggers

Primary Triggers


Valid Force Majeure 65 12.6 14.7 14.7Personal 103 20.0 23.3 38.0Electoral 175 34.0 39.6 77.6Mistake 34 6.6 7.7 85.3Party/Political 65 12.6 14.7 100.0Total 442 85.8 100.0

Missing System 73 14.2Total 515 100.0

Secondary Triggers


Valid Force Majeure 5 1.0 3.7 3.7Personal 16 3.1 11.9 15.6Electoral 45 8.7 33.3 48.9Mistake 12 2.3 8.9 57.8Party/Political 57 11.1 42.2 100.0Total 135 26.2 100.0

Missing System 380 73.8Total 515 100.0

(see Proposition 2), expecting that forced or duress exits (by definition, high-conflict affairs) would result in less successful successions than voluntary orinvoluntary exits. However, the dataset evidence for the conjecture is againweak and statistically insignificant.

These results suggest that whilst leaders who come to office as the resultof a high-conflict succession do face a moderately higher risk of beingdumped in the very short term, in general their survival prospects do notdeviate much from the mean. Only for in-government high-conflict suc-cessions, is the outlook really bleak (Blair to Brown comes to mind). For all other high-conflict successions, there is some element of risk as indi-cated by the higher proportion of very short-term leaders in the high-conflict group, however since conflict here seems to be impacting longevityrather than election results, it seems more likely that the risk to high-conflict successors comes from within their own parties.

Having gained office in a high-conflict succession requires the new leaderto be particularly competent at gaining the respect and soothing the woundsof his erstwhile internal opponents. Those who can not or will not are madeto pay. High-conflict leadership successions may well lower the threshold forfurther challenges, especially for non-governing parties. In the dataset thereare quite a few cases of ‘periods of instability’ where a number of shorter-termleaders with contested successions follow each other. The Australian Liberaland National parties cycled through a number of leaders each in the 1980sand early 1990s as they struggled to deal with the new role in opposition andtheir respective internal political conflicts came to dominate their operations.In the Israeli Labour Party the loss of government and the resignation offormer PM Ehud Barak ignited a wave of internal political struggles thatsaw four leaders in six years, before Barak himself returned in attempt torestore the party’s fortunes. Thus conflict can definitely beget conflict, andthis a risk that all parties face in the wake of high-conflict leadershipbattles.

However, our figures suggest that if a leader can ride out the wave ofparty conflict and hold onto power beyond the critical first years of a con-tested succession, then their chances for success are not greatly differentfrom their non-conflict successor counterparts. Conflicted successions donot necessarily lead to succession failure and voter backlash. Indeed, asStark (1996) also suggests, a change of leader, even if the preceding removaland selection process has been acrimonious, is just as likely to deliver arevival of the party’s fortunes. For example, despite the very public andoften bitter fashion in which they seized the leadership of their parties,Margaret Thatcher (UK-Conservative), Brian Mulroney (Canada-ProgressiveConservative), and Paul Keating (Australia-Labor) all presided over impres-sive electoral victories at post-succession elections. Voters may punish con-tinued disunity, but not necessary cathartic slaughter which produces aclear, undisputed new leader.


Conclusions: From explaining to assessing successions

Succession planning remains one of the most difficult and poorly executedfeatures of party procedure – a great deal of time and effort is invested intosmoothing that process as much as possible, yet even the most extensivelyorganized of them have had a tendency to go awry. In this chapter we haveattempted to shed some light on the complex and often problematic dynamicsof party leadership succession. We have sought to break some new ground inthe comparative study of party leadership succession. Combining literaturereview, hypothesis formulation, dataset construction and case study analysis,we tracked the behavioural correlates of succession outcomes.

Though explorative in our aims and methods, we have nonethelessidentified a number of potentially important and counterintuitive aspectsof the leadership succession process. For example, whilst we observe thatthe most popular choice for a successor is a ‘loyalist’ – someone with sup-portive views towards his or her predecessor and their party’s platform – their success in getting chosen does not translate into success with thepublic. Instead, those with clear reform agendas – the ‘reformers’ and ‘revo-lutionaries’ – seem to be the most successful types of leaders in terms oflongevity and election results. Similarly, whilst many would expect that‘disunity is death’ when it comes to internal leadership struggles, the factremains that most contested or high-conflict successions do not result inany particular public backlash. The highest risk for people who snatch theleadership from the incumbent rather than wait to have it bestowed uponthem in due course comes instead from within the party itself, where oldwounds never quite heal.

These findings allude to a remarkable divide between the needs anddesires of party and wider public: two different constituencies that a leadermust satisfy and yet seem to desire differing qualities from their leaders incertain contexts. As a result, political leaders engage in a delicate balancingact in order to survive and thrive. Though in some areas we suggest thatpublic popularity necessarily trumps positioning within the party, it becomesobvious that maintaining both is needed in order to effectively lead. Whilstconventional party wisdom might identify the perfect successor as one whostresses continuity and loyalty with the former leader, has strong and abidinglinks within the party, and succeeds his predecessor unchallenged, this studysuggests otherwise. It indicates that a maverick populist who emphasizes aclean break with party policy and takes down his predecessor in a party-roomshowdown might have just as much (indeed, if not more) chance of success-fully leading the party into the future.

The currently dominant institutional approaches to leadership successiondo not fully analyse this important choice, except in general terms. They shed insufficient light on the strategic choices the party elites face atthe impending exit of an incumbent leader. And they all but ignore the


complex interplay between party-internal and wider public actors and arenasin which judgements about incumbent, aspiring and newly emerged leadersare being made. A key step forward would be to modify and re-test the pro-positions and inferences provided here on a more elaborate data set, and ana-lyse more systematically than we have done here to what extent institutionalvariables at both the macro level (electoral and party system types) and themeso level (party rules and traditions of leader selection) mitigate the choicesof key protagonists in succession processes. In addition, further study of con-textual and temporal factors in successions is needed to better explain theresults found here.

Beyond the nitty-gritty of empirical theory-building, future research mayalso start to address the normative dimension of leadership succession: howdo we know a good one when we see one? Which criteria should be appliedto differentiate between successful, or at least less painful, cases from com-plete disasters? As a first stab at future research, we suggest that there arethree types of impact domains. First of all, what is the impact on policy? It isnot self-evident that leadership changes make a substantive difference forthe direction of party or government policy or lead to more effective partyand public policies (Bunce, 1981; Bunce and Roeder, 1986; Punnett, 1992;Boyne and Dahya, 2002). After all, many scholars argue that human factors inpolitics pale in comparison to the force of macro economic realities, complexinstitutions and trans-national mega trends. For those who beg to disagreewith this determinism, leadership transitions are a perfect opportunity to testtheir hypothesis. Can differences in policy preferences be detected betweenthe outgoing and the incoming leader? Does the new leader place new issueson the policy agenda? Does he put forward new solutions for old policy prob-lems? Does the new leader succeed in gaining political acceptance for hispolicy preferences? And if his policies are put into effect, does this lead to different and more strongly supported social outcomes?

Secondly, one can question how the popularity of outgoing and newlyincumbent leaders and their parties evolves before and after the leadershipchange. Ideally the new leader manages to gain so much support for him-self and his policies that he supersedes his predecessor, yet without com-pletely destroying his predecessor’s reputation. The newcomer must buildsufficient momentum without having to resort posthumous attacks on hispredecessor in order to gain stature by comparison. A new leader that doesnot succeed immediately in matching the popularity of his predecessor isnot necessarily politically dead, but he is handicapped. Even if he managesto gain policy successes and develops a distinctive political style he willhave to fight the increasingly mythical memory of his predecessor. Personalratings aside, the bottom line that many of the people keeping leaders inpower will apply to assessing the impact of a succession episode, is theparty’s position in the polls or – if they are wise – its de facto election resultsfollowing the conclusion of a succession episode.


The final criterion to assess the success of leadership transitions focusesupon the legitimacy of the regime. What does the leader succession do topublic trust in politics and to the legitimacy of the government? How docitizens value the rules of the political system before and after the suc-cession episode? It is plausible to expect that a sudden and prematuredeparture of an incumbent leader, or open warfare between incumbent andaspiring leaders not only do damage to the political stature of these peopleand their party, but also contribute to a decline of trust in politics as awhole. The legitimacy criterion is particularly relevant when leader suc-cessions coincide with changes in the ruling political coalition, or the polit-ical regime as a whole. Political regime changes have occurred frequently inLatin America and Eastern Europe. New leaders, some of whom were affiliatedclosely with the old regime, had to build support not just for themselves, butalso for the new, democratic political order. A comparison of some of thesecases suggests that the identity and style of individual leaders can in fact havea systemic effect: moral authorities as Mandela in South Africa and Havel inthe Czech Republic succeeded considerably better to strengthen democracy intheir countries than leaders who were survivors from the old, discreditedregime (Kane, 2001).

The three types of criteria are largely independent. This means that oneand the same case of leader succession can be judged a failure from oneperspective and at the same time a success from some of the others. Butthese criteria are not equally important in every situation of leader transi-tion. In the established democracies of Western Europe and North Americathe legitimacy of the political regime is so self-evident that it is hardlyuseful to evaluate leadership changes on this basis. It is more sensible tofocus on the popularity and policy success criteria. In new, less institution-alized political orders such as the Central and Eastern European demo-cracies, the ‘diffuse support’ (Easton, 1965) criterion becomes much morerelevant: new incumbents of high offices are not just expected to bringpeace and prosperity, they should also bestow authority on their office andtheir political system as a whole.

Whether we find any succession effects on policy, organizations and insti-tutions remains an open question. Some British research on the effects ofparty leadership changes suggests that it is wise not to overstate the expectedimpact of successions, and not to assume that because they are such risky,volatile occurrences for the people involved their institutional effects are morebound to be negative than positive (Punnett, 1992; Stark, 1996).

Notes

1 Australia, Austria, Bahamas, Barbados, Botswana, Canada, Denmark, Germany,Greece, India, Ireland, Israel, Jamaica, Japan, Luxemburg, Malta, the Netherlands,New Zealand, Norway, Portugal, Spain, Sweden, Trinidad and Tobago, and theUnited Kingdom.


2 Using newspapers, party histories, national political histories, almanacs and othersources; a continuous line of political party leaders was established for eachcountry for each of the major parties and a number of the larger minor parties.For each leader, a variety of variables were coded: age, sex, elections fought, elec-toral results whilst leader, time as leader, time as head of government (if at all),and so on. Additionally, for each leader a number of more interpretive variableswere coded corresponding to the nature of their transition out of power, includ-ing the type of exit (e.g. forced out, resigned), trigger for exit (e.g. illness, poorelection results), and attitude of successor to predecessor. Naturally for each ofthese interpretive variables, some judgement has been exercised on the part ofthe coder. However the parameters for the interpretive variables were clearlydefined, and in all cases the designation of these interpretive variables is beenwell supported by the source material. For unclear cases or cases with insufficientsource material, the coding has been left blank.

3 Due to the persistent presence of outliers amongst the categories, a 5 per cent trimmedmean is going to be referred to more commonly than the true mean as a way ofpresenting a more robust estimator.

4 This presents a baffling problem to students of succession: identifying and assess-ing ‘non-successions’. The problem is specifically that the category of ad hoc,endogenous triggers is difficult to pin down. When is party discontent with aleader strong enough to qualify as a genuine episode in which successionbecomes a real possibility? When are known rivals/revolutionaries among thefield of possible aspirants (or their supporters) really seriously trying to puttogether a take-over attempt? We can only know about cases that get into thepress and/or lead to formal party room or conference votes on the leadership.Beneath those, there is often a substantial, yet indeterminate number of instancesof covert agitation, polling and weighing of options that may not trigger formaldecisions to continue the incumbent but do serve to put pressure on the incumbent.Are these to be coded as ‘non-successions’ or are they merely pinpricks in a longerterm attrition strategy on the part of an aspirant and/or discontented party elites?

Bibliography

Aarts, K., Blais, A. and Schmitt, H. (eds) (2010) Political Leaders and DemocraticElections (Oxford: Oxford University Press).

Alderman, R. K. and Smith, M. J. (1990) ‘Can British prime ministers be given thepush by their parties?’, Parliamentary Affairs, 43(3), 260–72.

Andrews, J. and Jackman, R. (2008) ‘If winning isn’t everything, why do they keepscore? Consequences of electoral performance for party leaders’, British Journal ofPolitical Science, 38(4), 657–75.

Bennis, W. G. (1989) On Becoming a Leader (London: Century Business).Blondel, J. and Thibault, J.-L. (eds) (2009) Political Leadership, Parties and Citizens: The

Personalisation of Leadership (London: Routledge).Boyne, G. and Dahya, J. (2002) ‘Executive succession and the performance of public

organizations’, Public Administration, 80(1), 179–200.Bunce, V. (1981) Do New Leaders Make a Difference?: Executive Succession and Public

Policy Under Capitalism and Socialism (Princeton: Princeton University Press).Bunce, V. and Roeder, P. (1986) ‘The effects of leadership succession in the Soviet

Union’, American Political Science Review, 80(1), 215–24.


Cross, W. and Blais, A. (2008) ‘Who selects the party leader? A cross-national analysis’.Paper presented at the 2008 Elections, Public Opinion and Parties (EPOP) AnnualConference.

Davis, J. W. (1998) Leadership Selection in Six Western Democracies (Westport: GreenwoodPress).

Denham, A. and O’Hara, K. (2008) Democratizing Conservative Leadership Selection:From Grey Suits to Grass Roots (Manchester: Manchester University Press).

Easton, D. (1965) A Systems Analysis of Political Life (Chicago: Chicago UniversityPress).

Heppell, T. (2008) Choosing the Tory Leader: Conservative Party Leadership Elections fromHeath to Cameron (London: Taurus).

Heppell, T. (2010) Choosing the Labour Leader: Labour Party Leadership elections fromWilson to Brown (London: Taurus).

Kane, J. (2001) The Politics of Moral Capital (Cambridge: Cambridge 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.Kenig, O. (2009) ‘Democratization of party leadership selection: Do wider selec-

torates produce more competitive contests?’, Electoral Studies, 28(2), 240–7.King, A. (ed.) (2002) Personalities and the Outcomes of Democratic Elections (Oxford:

Oxford University Press).LeDuc, L. (2001) ‘Democratizing party leadership selection’, Party Politics, 7(3), 323–41.Lijphart, A. (1999) Patterns of Democracy (New Haven: Yale University Press).Marsh, M. (1993) ‘Selecting party leaders in the Republic of Ireland: Taking the lid

off party politics’, European Journal of Political Research, 24(3), 295–316.Panebianco, A. (1988) Political Parties: Organization and Power (Cambridge: Cambridge

University Press).Punnett, R. M. (1992) Selecting the Party Leader: Britain in Comparative Perspective

(London: Harvester Wheatsheaf).Quinn, T. (2005) ‘Leasehold or freehold? Leader eviction rules in the British Conserv-

ative and Labour parties’, Political Studies, 53(4), 793–815.Seldon, A. (1998) Major: A Political Life (London: Weidenfeld and Nicolson).Stark, L. (1996) Choosing a Leader: Party Leadership Contests in Britain from Macmillan

to Blair (London: Macmillan).Turner, J., Subasic, E. and Reynolds, K. (2008) ‘Identity confers power: The new view of

leadership in social psychology’, in P. ’t Hart and J. Uhr (eds) Public Leadership:Perspectives and Practices (Canberra: ANU E Press), pp. 57–72.

Weller, P. (1983) ‘The vulnerability of prime ministers: A comparative perspective’,Parliamentary Affairs, 36(1), 96–117.

Weller, P. (1985) First Among Equals (Sydney: Allen & Unwin).Weller, P. (1994) ‘Party rules and the dismissal of prime ministers: Comparative

perspectives from Britain, Canada and Australia’, Parliamentary Affairs, 47(2),133–43.


8Holding Party Leaders to Account:The Westminster Cases*William Cross and André Blais

Introduction

Political party leaders have long been recognized as central figures inWestminster parliamentary democracies. In the legislative arena they fillthe leading roles of Prime Minister and Leader of the Official Opposition(Weller, 1985). Parties hold a privileged place in parliamentary electionsand their leaders are the central figures in these campaigns (McAllister,1996). Similarly, leaders exercise considerable control over important intraparty decision-making (Heidar and Koole, 2000). While leaders have alwaysplayed a central role in modern parliamentary democracies, observers haveidentified an increase in their authority in all three realms – legislative,electoral and intra party (Poguntke and Webb, 2005).

Given this key role of party leaders it is not surprising that in recent yearsscholars have paid close attention to the manner in which leaders areselected. Many have observed an increase in the influence attributed torank-and-file party members at the expense of party elites and particularlythe parliamentary party (Cross, 1996; Carty and Blake, 1999; LeDuc, 2001;Kenig, 2009; Cross and Blais, forthcoming). Coupled with this observationof ‘democratization’ of leadership selection through expansion of the selec-torate has been the charge that this reform works to increase the authorityof the leader. This argument, made by Mair (1994), is that diffusing theauthority to select leaders to a larger, less proximate and more disparatebody than a party’s parliamentary group makes it more difficult for anyoneto hold leaders accountable and potentially to remove them from office.Beyond these general characterizations, and with few notable exceptions(Weller, 1994; Courtney, 1995; Quinn, 2005; Bynander and ’t Hart, 2007),little has been written focusing on the ability of parties to hold their leadersto account and specifically to remove them.

In this chapter we consider the ways in which contemporary politicalparties are able to hold their leaders accountable and whether and howthey can remove them. In doing so we argue that party rules matter

133

insofar as they influence the relative security of a party leader, determinewho he or she is accountable to, and thus affect their relations with theirparty colleagues.

We present a typology of party leadership accountability that is based on three characteristics that define a party’s approach to leadership reviewand removal. The first variable is whether a party has a formal mechanismthat allows it to remove a no longer wanted leader. The second and thirdconcern the ease with which a party can remove a leader by consideringwho has the removal authority and when that authority can be exercised.

Our cases include parties from the five principal Westminster countries:Australia, Canada, the Republic of Ireland, New Zealand and the UnitedKingdom. Because we are interested in the balance of authority betweenthe parliamentary and extra parliamentary parties we restrict our analysisto the 22 parties that finished in the top five, in terms of seats won, ineither of the two most recent elections to the national lower house (as of31 December 2007). In the section considering leadership transitions weinclude four additional parties that have merged to form a new party that iscaptured in our sample.1 Thus, we examine both parties that are in and outof government and include in our sample leaders who are serving as Prime Minister and others who are in opposition. As illustrated below,party rules, along with the broader mandate they enjoy, typically, thoughnot always, provide prime ministers with more security in their position asparty leader.

In the Westminster systems it is easy to identify a party leader, as there isunity between the leadership of the parliamentary and organizational, orextra parliamentary, party. In all parties, the parliamentary leader is theuncontested head of the party and maintains authority both in electioncampaigns and in the legislature, the only exceptions being the New ZealandGreen and Maori parties, with male and female co-leaders sharing respons-ibility for the parliamentary and extra-parliamentary wings.

Both secondary and primary data sources are used. We have systemat-ically reviewed both contemporary and historic party statutes, newspaperaccounts of leadership transitions, and the literature on parties and leaders’biographies. We have also conducted a series of approximately 40 semi-structured elite interviews, in all five countries, with party officials, acad-emic experts and political journalists. Current and former party leaders,members of parliament, party secretaries and senior party staff were amongthose interviewed.2 In order to preserve confidentiality individual inter-viewees are not identified.

Reasons for leader exits

We begin our analysis with an examination of the 110 cases of leadershiptransition in 26 parties occurring between 1965 and 2007. We consider

134 Holding Party Leaders to Account: The Westminster Cases

the circumstances under which these leaders left office and classify eachdeparture of a leader into one of six categories: resigned voluntarily; resignedunder pressure; resigned because of a single, specific disagreement with theparty; leadership ended because the party no longer exists (typically a mergercreating a new party); death of the incumbent; and the incumbent beingformally removed.

It is easy to assign cases to each of the last four categories and to identifythose that belong in one of the first two. Significant difficulty occasionallyarises in determining when a resignation is voluntary and when it resultsfrom pressure placed on the leader to leave. In both instances the resign-ation is voluntary, as the leader is not compelled to resign. Nonetheless, it is useful to differentiate between cases when the leader essentially retiresbecause of a wish not to continue and those cases in which a leader quitslargely because of a realization that there is significant opposition to theircarrying on as leader and that they are likely to face a formal leadershipchallenge that they may not win. There is always a colleague who fanciesthe leadership and would prefer the incumbent leave to open up a shot atthe top job. When this pressure reaches the point of pushing the incum-bent out is a judgement call that we are forced to make, and in doing so wedraw upon interviews with country experts. It is important to bear in mindthat the number of cases where this is a real issue is limited and thus dis-cretion is restricted to a small number of cases. If we have erred it is in thedirection of categorizing cases as the leader leaving under his/her own discretion despite being under some party pressure to do so.

The most compelling finding is that leaders rarely leave willingly. Of our110 cases, only 26 can be classified as retiring voluntarily. In 55 cases theleader resigned under significant pressure and in 17 others the leader wasformally removed. The remaining cases include four deaths, five partiesthat merged with others thus ending the incumbent’s tenure, and threecases where leaders resigned because of a specific, fundamental differencewith their party often over an organizational question.

In 20 cases the leader was prime minister at the time of departure. Theseleaders are significantly less likely to be ‘forced’ from the job than are theiropposition colleagues. While two-thirds of opposition leaders were either formally removed or resigned under significant pressure, the same is true for half of retiring prime ministers (11 of 20). We find three cases of sittingprime ministers formally removed from their party leadership: Robert Hawke(Australia Labor), John Gorton (Australia Liberals) and Margaret Thatcher (UK Conservatives).3 Eight other sitting Prime Ministers resigned in the face ofsignificant party opposition to their continuing as leader, and several of thesefaced a high likelihood that they would lose a forthcoming confidence vote.These include Jim Bolger (New Zealand National), Charles Haughey (Ireland’sFianna Fáil), Jack Lynch (Ireland’s Fianna Fáil), Albert Reynolds (Ireland’sFianna Fáil), David Lange (New Zealand Labour), Geoffrey Palmer (NZ Labour),

William Cross and André Blais 135

Jean Chrétien (Canadian Liberal) and Tony Blair (UK Labour). Two others diedin office and seven retired voluntarily.

While there are cases of leaders being formally removed in all countries,the incidence of this varies dramatically. In both Canada and Ireland onlyone party leader was formally removed during the more than four decadesstudied (John Diefenbaker from the Canadian Progressive Conservatives andJohn Bruton from Ireland’s Fine Gael) and three in the United Kingdom (allin the Conservative Party: Edward Heath, Thatcher and Ian Duncan Smith).This represents less than 5 per cent of Canadian and Irish leadership transi-tions and slightly less than one in seven in the British case. This can be con-trasted with the Australian and New Zealand cases. In these countries, morethan one-quarter of leaders have been formally removed from office. InAustralia this is the case for seven of 28 leaders and in New Zealand for five of 17. All three of the Australian parties in our sample have formallyremoved leaders while in New Zealand this has been restricted to the two largest parties: National and Labour. Two of the three ousted primeministers led Australian parties.

Given the findings that leaders rarely leave fully of their own accord, andthe difference in exit patterns among countries with generally similar, thoughnot identical, institutional contexts, it is important to consider the rules sur-rounding leadership removal. As Weller (1994: 133) suggests in the contextof prime ministerial dismissal, ‘rules can determine politics’. We find thatparties take a wide range of approaches relating to leadership accountabilityand removal. We focus on three questions: does a mechanism exist to for-mally remove a leader? If so, how can the leader be removed? And, finally,who has the authority to remove the leader? The answers to these questionssignificantly impact on a leader’s security in office and influence his/herrelationship with both the parliamentary caucus and extra parliamentarymembership. As Weller (1983: 99) suggests in his examination of the vulner-ability of prime ministers to internal revolt in the United Kingdom, Canadaand Australia, their ‘similar parliamentary system, similar constitutionalassumptions, similar cabinet principles and prime ministers with similar …powers’ make it ‘easier to see how procedures and institutions contribute todegrees of vulnerability’.

Can party leaders be removed?

Beginning in the mid-1960s many parties began adopting formal rules gov-erning leadership selection and, either at the same time or shortly there-after, many adopted rules and norms for the removal of leaders. Prior tothis period, there often existed no formal mechanism for the review andremoval of the leader. In some it was assumed that the parliamentarycaucus could remove the leader since they had chosen him, though thisauthority was not widely used and in many parties there was an assump-


tion that leaders could not be forced from office (see Weller, 1983: 97).More often, caucus and extra parliamentary party pressure was brought toencourage a leader to leave. This was the case particularly in parties wherethere were no formal rules for leadership selection and thus no group thatcould claim legitimate authority to remove the leader. An example of this is the British Conservative party in which until 1965 leaders were said toemerge through the party’s ‘magic circle’ which essentially meant secre-tive consultation among a few party elites, and, when the party was in government, the Crown (Alderman, 1999). As the parliamentary caucus didnot have the authority to choose a leader it could not legitimately claim removal authority. This situation changed when the party moved toleadership selection via a formal vote of the parliamentary caucus whichthen gave the same body legitimacy in claiming the right to remove theleader.

The Canadian case provides examples of parties with formal selectionrules that did not have any provision for removal. Leaders in Canada havebeen chosen by the extra parliamentary party since 1919 (Courtney, 1995),yet no removal mechanisms existed until the late 1960s. As late as the mid-1960s, Liberal leader Lester Pearson correctly argued that ‘the leader deposeshimself,’ as no other group has the authority to remove him (Engelman and Schwartz, 1975: 243). In the case of the Progressive Conservatives this led to a protracted standoff in which the extra parliamentary partyclaimed the removal authority and the incumbent leader, John Diefenbaker,argued that they did not have this power. The extra parliamentary partyultimately prevailed, removing Diefenbaker in 1967 and in the processclaiming for itself the power to remove future leaders.

We find that all of our parties now incorporate some formal process thatallows for the removal of an unwanted leader. However, in two cases, theCanadian Liberals and Conservatives, this ability only exists when theparty is in opposition. In both parties, when the leader is serving as PrimeMinister, there is no formal ability for either the parliamentary or the extraparliamentary party to remove him/her. This is a meaningful exception asone or the other of these parties’ leader has served as prime minister for thecountry’s entire history. In the other 20 parties leaders can be removedboth while in government and in opposition, though several make removalof a prime minister more difficult than that of an opposition leader. Primeministers have often argued that their mandate comes from general elec-tion voters and thus they should not be removable by an intra party revolt(for example Thatcher in 1990 and Australia’s Rudd in 2010). While we stillfind some degree of deference towards prime ministers it is only in the twoparties mentioned above that they are exempt from the possibility offormal removal. Other parties, such as Labour and the Liberal Democrats inthe UK and Fine Gael in Ireland, make it somewhat more difficult to removea party leader serving as PM (or Taoiseach) but do allow for the possibility.


How can leaders be removed?

We find three general ways for leaders to be removed from their position:expiry of a fixed term, failure to win a prescribed confidence vote, and thepassing of a spontaneous motion to remove the leader. Each of these isconsidered below.

Fixed leadership terms

One of the key characteristics distinguishing democracies from other polit-ical systems is that leaders do not serve indefinitely but, rather, periodicallyrequire endorsement from their voters in order to maintain their position.Powell (1982: 3) has suggested that one of the necessary conditions fordemocracy is that ‘leaders are elected at regular intervals’, and Massicotteand Blais (2000: 308) suggest that ‘all elections are for a specific term’.Given this, we are surprised to find that a majority of parties continue toselect their leaders for indefinite terms. As illustrated in Table 8.1, in fewerthan half of our cases (nine of 22) do leadership terms automatically expire.In none of these cases is the leader prohibited from seeking additionalterms and in many the norm is for the leader to be re-elected.

When there are fixed leadership terms they range in length from one tofive years. For example, in the UK Labour party, when in opposition, lead-ership elections take place annually at the time of each party conference.Would be challengers to the incumbent need only collect a requisite num-ber of caucus signatures in order to trigger a full leadership contest via theparty’s electoral college mechanism. Similarly, when the New Zealand Maoriparty was formed in 2005, its rules required leadership contests at eachannual party ‘hui’ or conference (this rule was subsequently changed sothat elections are now scheduled to be held once every three years). At theouter limit, in terms of length of tenure, are leaders of the Irish LabourParty, who serve five year terms with the ability to stand for re-election.

Several parties, particularly those in which the parliamentary party choosesthe leader, tie the term of office to the electoral cycle. This is the case inmany Australian and New Zealand parties. For example in the New ZealandLabour party, automatic leadership elections are held at the first caucusmeeting in the year prior to a scheduled election. Others, includingAustralia’s Labor, Liberal and National parties, hold leadership contests at thefirst meeting of the parliamentary party following each general election,essentially attaching a three year term to the leadership position with thepossibility for re-election.

While the mechanism and timing varies among these parties, the pres-ence of a fixed term ensures that at some point a party can rid itself of anunwanted leader simply by not re-electing him/her. Specified terms canhave two important effects – one potentially increasing the security of anincumbent leader and the second somewhat negating this effect. First, they



Table 8.1 Methods by which party leaders may be removed (N = 22)

Automatic Spontaneous Fixed term periodic review review

AustraliaLiberals 3 years* XNational 3 years* XLabor 3 years* X

New ZealandLabour 3 years* XNational X XACT XGreen XMaori 3 yearsNZ First XUnited Future X

United KindomLabour (oppos) 1 year

(gov’t) +Conservatives XLib Dems (oppos) 5 years* X

(gov’t) X

IrelandLabour 5 years XFianna Fáil XFine Gael (oppos) X X

(gov’t) XProg. Dems X XGreens 5 years* X

CanadaLiberals (oppos) X

(gov’t)Conservatives (oppos) X

(gov’t)New Democrats XBloc Québécois X

Totals 9 8 15

(for each column total a party is counted once, not separately for when in opposition andgovernment.)*these are tied to the electoral cycle, thus the term may be shortened in the event of an earlyelection.+ the UK Labour party, when in government, does not provide for an automatic review vote,but as described above provides annual possibility for a leadership contest to be launched.

can provide incumbent leaders with security during the course of theirterm, as they sometimes argue that those wishing to see them removedshould wait until the end of the term and raise their challenge then. Thesecond effect is that the presence of regularly scheduled contests provideslegitimacy for a challenge at the end of the leadership term without theappearance of an ‘insurrection’ against an incumbent.

The first effect is also diminished by the fact that, as described below, theexistence of a fixed term typically does not prevent the bringing of a spon-taneous challenge. Thus while the leader may claim the moral authority tolead during his term in office, mechanisms exist, and are often used, toremove him/her before the term ends. This is particularly the case whenthe parliamentary party holds the removal power.

Scheduled confidence votes

Many parties without fixed terms require that their leaders periodicallyreceive a vote of confidence in order to remain in their position. These varyin the form of the question asked from whether there is confidence in thecurrent leader to whether there is desire for a new leadership contest. Theresult is the same in either case – a leader losing such a vote must eitherstand in a new election or leave the leadership position. These cases aresimilar to fixed terms in that the leader cannot stay in office without anindication of support from their electorate.

None of the parties with fixed terms require that leaders face scheduledconfidence votes during their prescribed term. Eight of the other 13 partiesdo require them. While similar in effect, these differ from fixed terms inthat there is not a contest and thus there is no need for someone to stepforward and offer a challenge to the incumbent. The vote is a straightthumbs up or down on the current leader.

In many parties these opportunities for removal votes are held once perelectoral cycle, often at the first opportunity following a general election.This is the case in New Zealand National, Fine Gael and the CanadianLiberal and Conservative parties when in opposition. The review takes placemore frequently (every one or two years) in the Canadian New Democratsand Bloc Québécois and New Zealand Greens as leadership confidence votesin these parties are automatically on the agenda at every party conference.

As illustrated in Table 8.1, there are five parties in which leaders do notrequire an indication of support in order to remain in office (either throughre-election or winning a confidence vote). This category includes both smallparties and large governing ones: Fianna Fáil in Ireland, the Conservatives inthe United Kingdom, and, in New Zealand, United Future, ACT and NewZealand First.

Five other parties have different rules depending upon whether they arein government or opposition. In each case these parties have attempted tomake their leaders more secure while in government by either eliminating


the ability to remove them or making it more difficult to do so. Fine Gaelleaders, when in opposition, are subject to a confidence vote followingeach general election while no such automatic vote is held when the partyis in government. In UK Labour, when in opposition, leaders are subject tore-election at the time of each annual conference. In government, removalof leaders is much more difficult. While they can be removed, the bar is sethigh requiring both the signed support of 20 per cent of MPs for an alter-native candidate and a majority vote of annual conference delegates sup-porting a new leadership election. The UK Liberal Democrats proviso for a five year fixed term for leaders includes the caveat that a term will not expire while the party is serving in government. Thus, including the Canadian Liberals and Conservatives, we find that in ten of 22 cases,leaders (sometimes only while in government) can serve indefinitelywithout need of any periodic endorsement from their party.

Leadership politics in parties with regular review votes have similardynamics to those with fixed terms in that leaders often argue that challenges brought at other times should be put off until the scheduledreview vote, and disgruntled party members are guaranteed an opportunityto express their displeasure with the leader. We do find that when thereview authority lies with the parliamentary party regularly scheduledreviews do not limit their ability, or willingness, to bring spontaneous challenges.

Spontaneous leadership challenges

The third removal method is the ability for a group within the party tobring a spontaneous challenge to the leader. Typically, this categoryincludes cases of the parliamentary party being able to bring either a con-fidence or ‘spill’ motion against an incumbent leader at any time. A spillmotion essentially declares the leadership vacant and, if passed, results in anew leadership contest. As illustrated in Table 8.1, the existence of a fixedterm in office, or of a regularly scheduled review vote, does not necessarilyprevent efforts to remove a leader at an earlier time.

We find that in a majority of our cases, 15 of 22, leaders are subject toremoval at any time. Most cases of leaders being removed by their partyhave occurred through a spontaneous challenge – even when the rulesprovide for fixed terms and regular review votes. In most of these parties,the possibility of spontaneous removal is through a vote of the parliamen-tary party. In all but one case in which the parliamentary party has theremoval authority it is not restrained in terms of when it can exercise it.The one exception is Fine Gael in which a leader who survives a confidencemotion is immune from further challenge for a period of six months. Theintent of this provision is to provide some stability within the party. Partiesallowing for spontaneous removal generally do not distinguish betweenwhether or not the party is in government.


Spontaneous challenges often materialize very quickly. We find cases ofincumbent leaders having no knowledge of a brewing insurrection untilthe confidence motion is brought in the party room or until just prior tothe caucus meeting. Surprise is often a desired tactic of the rebels as itdenies the incumbent opportunity to organize and rally supporters (see dis-cussion below for an example).

Parties that have vested the removal authority more clearly with theextra parliamentary party, typically in the party conference, provide oppor-tunities for removal only at limited, prescribed times. This is because thevote generally takes place at the party conference which is set for a speci-fied time. The only exceptions are two now defunct Canadian parties, theAlliance and the Progressive Conservatives, in which the extra parliamentaryparty had the ability to bring forward the leadership review vote in advance of a scheduled party conference. In both cases the threshold required to do so was an indication of support for an early vote from a significant number of constituency party associations. Even in these cases, however, it would inall likelihood have taken a minimum of several months for the party to organ-ize the vote among the extra parliamentary members (the provision was never successfully used in either party). This can be contrasted with the ability of MPs to bring a non-confidence motion and have it voted on either immediately or within days.

In cases where the extra parliamentary party has the removal authority,one result can be leadership struggles playing out over long periods of timeas leaders face revolt from within the party but their opponents have noimmediate opportunity to remove them. A common result is that embat-tled leaders hang on in the face of widespread opposition in the parlia-mentary, and sometimes rank-and-file, membership knowing that theycannot be formally removed until a set date. In the end, these leaders tend toremain in the leadership until the review opportunity is imminent beforeannouncing their retirement plans (see discussion below for examples).

Another important difference in method is whether an alternative leadermust be named prior to launching an effort to remove an incumbent. Thisdifference is manifested in rules that either allow for a discrete review orremoval of the incumbent versus those that require dissenters to spark a fullblown leadership contest. The effect of fixed terms and spontaneous chal-lenges essentially is to require a challenger to emerge to confront the incum-bent, whether in a party room vote or a universal membership ballot. Wecan contrast these with automatic review votes that simply ask whether theleader maintains the confidence of those voting. If the leader has lostconfidence then a full contest ensues. ‘Spill’ motions brought in the partyroom are similar in that they do not require the naming of an alternativeleader. These mechanisms do not require that a caucus colleague formallystand against the incumbent and in doing so risk career retribution shouldthe challenge fail.


Who can remove the leader?

Today, formal involvement of the extra parliamentary party in leadershippolitics is becoming the norm. While the parliamentary caucus selectedleaders in the vast majority of our parties well into the latter half of the20th century, they retain the full authority to do so in only eight of our22 cases, which are mostly found in Australia and New Zealand (Cross andBlais, forthcoming). We do not, however, find a shift of the same magnitude in terms of who has the authority to remove the leader. Werestrict our analysis to which party group has the authority to cause an


Table 8.2 Who can remove the leader?

Either theparl.

Parl. Party Party caucus or Party caucus conference committee constituencies member

AustraliaLiberals XNationals XLabor X

New ZealandLabour XNational XACT X <both X

needed>Green XNZ First XUnited Future X

United KindomLabour (gov’t) X <both X

needed>(oppos) +

Conservatives XLib Dems X

IrelandLabour XFianna Fáil XFine Gael XProg. Dems XGreens X <both X

needed>

CanadaLiberals (oppos) XConservatives (oppos) XNew Democrats XBloc Québécois XTotals 11 4 1 1 1

exclusive exclusive exclusive exclusive exclusive2 shared 1 shared 2 shared 1 shared

(+When UK Labour is in opposition, there is an automatic opportunity for a contest at each annual conferencewhich is triggered by a challenger reaching a required threshold of support from the parliamentary caucus(essentially a requirement of candidacy); there is, however, no opportunity to remove the leader betweenconferences).

unscheduled leadership contest, either by voting no confidence in anincumbent leader or voting in the affirmative when asked if a leadershipcontest should be held, in either a scheduled or spontaneous review vote.All but one of our parties provide for the possibility of this occurring. Theonly exception is in the NZ Maori party where there are no rules allowingfor removal of leaders during their three year term (as discussed above, insome other parties the authority only exists when in opposition). Table 8.2indicates who has the authority to remove an incumbent leader in eachparty.

In slightly more than half of our cases (11 of 21) the situation is clear asthe parliamentary caucus retains the sole and unencumbered authority toremove the leader. In these cases, there is no formal role for the extraparliamentary party in deciding whether a leader remains in the job. Thiscategory includes three parties in which rank-and-file party members par-ticipate in the selection of leaders. While the UK Conservatives andIreland’s Fine Gael and Progressive Democrats provide a significant role formembers in leadership selection, there is no parallel authority for removal.4

This potentially creates a situation of conflict between the parliamentaryparty and the grassroots membership as MPs can vote out a leader chosenby the party’s members. We find one case of this occurring with the UKConservative membership choosing Ian Duncan Smith in 2001 who wassubsequently removed by the parliamentary party in 2003 before contest-ing a single general election. In one additional party, the UK LiberalDemocrats, the party caucus can remove the leader as can the extra parlia-mentary party, and each is able to do so without the cooperation of theother. At the other end of the spectrum, we find only one case of a partythat provides for a ballot of its members to be held for a leadershipconfidence or removal vote – the Canadian Liberal party, and this onlyoccurs after elections in which the party does not form the government.

In the remaining parties, the removal authority rests with either theparty conference or some combination of the parliamentary and extra par-liamentary parties. Party conferences have full and sole removal authorityin four parties: the Canadian Conservative, Bloc Québécois and NewDemocratic parties and the New Zealand Greens. In three parties removalof the party leader requires action of more than one branch of the party. In UK Labour both the parliamentary party and the conference must act to spark an unscheduled leadership contest (the opportunity only exists in government). In the Irish Green Party a leader can be removed by theNational Executive Committee only if it receives a request to do so fromeither 12 local constituency groups or half of the party’s TDs and localauthority members. In New Zealand ACT, only the party’s ManagementBoard can remove the leader. However, it can only act after receiving arequest to do so from a majority of the parliamentary caucus. In Ireland’sLabour party, the party’s Central Council (comprised of representatives


from each constituency party and the party’s executive committee) can, bya two-thirds vote, remove the leader.

Thus, while the parliamentary caucus acting on its own can remove theleader in 12 of 21 cases, only in six of the 21 is it necessary for there to be asignal of support from the party’s grassroots members either through a voteat party conference, support from constituency associations or a ballot ofmembers. This is the case for the New Zealand Green, UK Labour and thefour Canadian parties. In the remaining three cases (Irish Greens andLabour, and New Zealand ACT) the rules provide the possibility for leadersto be removed without the membership party being involved throughvarious combinations of elected officials and central party elites workingtogether.

It is important to note that even in cases in which there is no formal roleascribed to the extra parliamentary parties members often do have influencein leadership politics. Members of the parliamentary party depend on thesupport of local party activists for reselection as candidates for the nextelection, and for the financial and volunteer support necessary to mount avigorous local campaign. While this varies somewhat depending on thetype of electoral system in place and the degree of decentralization in can-didate selection, members of the caucus in all countries reported in inter-views that they consult with their local activists in making these decisions.In several cases MPs recounted being under significant pressure from theirlocal supporters to take a particular position in a leadership struggle, oftenwith the threat of a reselection fight and or withholding of campaignresources if they did not act accordingly. Two high profile cases of the extraparliamentary party exercising influence on the parliamentary party’sefforts to remove leaders are those of Charles Haughey in Fianna Fáil andBill Rowling in New Zealand Labour.

It is well documented that Rowling’s leadership of the New ZealandLabour party in 1980 was saved by pressure from the extra parliamentaryparty. Facing a challenge in the party room, Rowling appeared headed tocertain defeat in the caucus vote before the extra parliamentary party wasactivated on his behalf. The party president, Jim Anderton, fought a spir-ited battle for Rowling. Faxes and phone calls from constituency activistsfilled the offices of targeted MPs, some of whom feared for their reselection.Many observers credit the organizing of the extra parliamentary party withtipping the balance in Rowling’s one vote victory (see, for example,Henderson, 1981: 25).

Party activists were also very involved in the leadership struggles sur-rounding Charles Haughey in Fianna Fáil. While the party rules provided norole for the extra parliamentary membership, Haughey worked endlessly tocultivate support at the constituency level and encouraged party membersto contact TDs and express their support for him. When faced with a leader-ship challenge, Haughey’s supporters in the constituencies were known to


organize phone calls to TDs on his behalf and on occasion to travel toDublin and march in front of Leinster House. While many in the partyobjected to what they viewed as outsider interference in the work of the par-liamentary party, Haughey openly encouraged the involvement of theseactivists and regularly insisted on open votes in the party room so that TDscould be held accountable by their local party members. The leadershipstruggles during Haughey’s tenure were quite rancorous and pressures onTDs from outside the party caucus were often direct and occasionally threat-ening. One former senior cabinet member recounted an occasion in whichthe brother of a TD received phone calls from four of his firm’s five largestcustomers promising to cut off business with him if his brother did notsupport Haughey. On another occasion, a riled up activist assaulted a TD,who had opposed Haughey, outside the Dail.

Similarly, when the removal authority lies with the extra parliamentaryparty, this does not absolve leaders from maintaining support from theirparliamentary colleagues. This has been most apparent in the Canadiancases in which removal authority has been vested in the extra parlia-mentary party for several decades. While embattled leaders have regularlyargued that their parliamentary colleagues lack the authority to removethem from office, leaders lacking support in the party room inevitablysuccumb to such pressure, as an unsupportive parliamentary party is able to make their life as leader unpalatable. An illustrative example is that of Canadian Alliance leader Stockwell Day. After leading his party to a dis-appointing electoral finish in 2000, Day was under significant pressurefrom his parliamentary colleagues to relinquish the leadership. He arguedthat he could only be removed by the extra parliamentary party at a partyconference to be held in approximately 15 months’ time. Significant seg-ments of the parliamentary party revolted against Day, with more than adozen members leaving the parliamentary caucus. Eventually, Day’s abilityto lead the party was so diminished that he faced no real option to relin-quishing his leadership even though formally he could not be removedexcept by party convention. Leaders who can formally be removed by theextra parliamentary party may be able to hang on for some months in theface of significant opposition from the parliamentary party but are unlikelyto be able to maintain the leadership in the long run without support fromtheir elected colleagues.

Comparing and interpreting party leader accountability

What is clear from the above discussion is that there is no single norm interms of how leaders are held accountable and potentially removed. Thisleaves the question of how best to make sense of these many differentapproaches and what their significance is in terms of a leader’s security inoffice. Bynander and ’t Hart suggest that: ‘Different rules for party leader


(de)selection do produce significant differences in the vulnerability of leadersto removal attempts’ (2007: 49). As displayed in Table 8.3, we suggest that the key variables in this regard are: whether a removal mechanism exists;whether the extra parliamentary party must in some way act to remove theleader; and when the parliamentary party has this authority, whether it is inany significant way encumbered in its use. The most secure leaders are thosewho cannot be formally removed. While pressure may be brought to try andforce them from the leadership, at the end of the day, the decision and timingis ultimately theirs alone. These are closely followed by those whose removalrequires the participation of the extra parliamentary party.

We find that this authority is almost never formally used. While the threatof removal by the extra parliamentary party has resulted in leaders leavingoffice, the difficulty in organizing this large and dispersed group, and the timeinvolved in doing so, provides leaders with significantly more security thanwhen the parliamentary caucus can depose them. Those removable by theirparliamentary colleagues are the most vulnerable. We divide these cases intotwo categories: those in which the parliamentary group’s ability to fire theleader is essentially unchecked and those in which this authority is in someway limited. The latter category primarily consists of instances where the parliamentary party does not select the leader on its own but rather shares


Table 8.3 Relative security of party leaders

Most secure Least secure

No possibility removal requires parliamentary party parliamentary of removal action by extra removes with some party removes

parliamentary impediments unemcumberedparty

(Gov’t) Cdn. (oppos) Cdn. NZ ACT Aus. NationalsConservatives Conservatives

Irish Fine Gael Aus. Liberals

(Gov’t) Cdn. (oppos) Cdn. UK Tories Aus. LaborLiberals Liberals

Cdn. NDP (oppos) UK Labour NZ Labour

Cdn. Bloc UK Liberal Democrats NZ National

(gov’t) UK Labour Irish Progressive Dems. NZ First

NZ Greens NZ United Future

Irish Labour Irish Fianna Fáil

Irish Green

NZ Maori

this responsibility with the extra parliamentary party. We suggest that anexpanded selectorate tempers a parliamentary party’s use of the removal pre-rogative and we find that the vast majority of removals have occurred inparties where the parliamentary caucus retains both the selection and removalauthority.

When we consider the four categories suggested in Table 8.3, we see thatonly one leader has been removed in a case in which the extra parliamen-tary party holds the removal authority, while 16 have been removed by theparliamentary party. The 1967 case of Canadian Conservative leader Diefen-baker remains the only instance of a leader being ousted by the extra parlia-mentary party. Of the other 16, there is only one case in which the authorityof the parliamentary party in leadership politics was in any real sense limited.In 15 of the cases leaders were selected by the party room and subject toremoval by them at any time.5

The data in Table 8.4 support the argument made by Mair (1994) andothers that expanding the leadership selectorate provides greater securityand makes a leader less likely to be formally removed. This is true for severalreasons. First, we find that when the power to select the leader is expandedbeyond the parliamentary party, MPs are often stripped of the removalpower. When a wider institution, typically the party conference, is vested


Table 8.4 Leaders formally removed (1965–2007)

Australia (7)John Gorton, 1968 (Liberal) parliamentary party unecumbered government loss seatsBill Snedden, 1972 (Liberal) parliamentary party unecumbered opposition held steadyJohn Howard, 1989 (Liberal) parliamentary party unecumbered opposition held steadyIan Sinclair, 1989 (Nationals) parliamentary party unecumbered opposition loss seatsRobert Hawke, 1991 (Labour) parliamentary party unecumbered government loss seatsJohn Hewson, 1994 (Liberal) parliamentary party unecumbered opposition loss seatsKim Beazley, 2006 (Labour) parliamentary party unecumbered opposition *

Canada (1)John Diefenbaker, 1967 (Cons) extra parliamentary party opposition held steady

Ireland (1)John Bruton, 2001 (Fine Gael) parliamentary party unecumbered opposition gained seats

New Zealand (5)Arnold Nordmeyer, 1965 (Lab) parliamentary party unecumbered opposition held steadyRobert Muldoon, 1984 (Nat) parliamentary party unecumbered opposition loss seatsJim McLay, 1986 (Nat) parliamentary party unecumbered opposition loss seatsMichael Moore, 1993 (Lab) parliamentary party unecumbered opposition gained seatsBill English, 2003 (National) parliamentary party unecumbered opposition loss seats

United Kingdom (3)Edward Heath, 1975 (Cons) parliamentary party unecumbered opposition loss seatsMargaret Thatcher, 1990 (Cons) parliamentary party unecumbered government loss seatsIan Duncan Smith, 2003 (Cons) parliamentary party ecumbered opposition *

(* Beazley and Duncan Smith did not lead their party in the election prior to their removal).

with the review and removal authority the threat is less proximate as it isonly able to be exercised at prescribed times – typically once per electoralcycle. Extra parliamentary party members are also not as vested in day today party affairs and have less at stake in the party’s short-term success thando MPs. The result is that they both have less opportunity and may be lessinclined to remove an incumbent who they selected as leader. When theparliamentary group maintains the removal authority, it appears less likelyto use it when there is a broader selectorate. Elected officials we interviewedsuggested that parliamentary parties are reluctant to remove a leader chosenby a broader selectorate, and indeed we have only one case of this occurring– the removal of Ian Duncan Smith by the UK Conservatives.

The relative security of the leader influences his/her reaction to an immi-nent challenge and the character of any transition to a new leader.Challenges brought against leaders chosen by their parliamentary col-leagues tend to be resolved quickly, either through an immediate partyroom vote or the sudden resignation of a leader facing defeat in such avote. We can contrast these with long, drawn out struggles in parties inwhich the extra parliamentary party is formally involved in the removalprocess. There are many examples of this with the most recent at the timeof writing being that of Gordon Brown in the United Kingdom. Labourparty rules provide that their leader, when in government, can only beremoved if a petition in support of another candidate as leader is endorsedby one-fifth of MPs and a subsequent motion calling for a new leadershipcontest is passed by a majority of delegates at the annual party conference.These rules made Brown very difficult to remove. A challenge would haverequired a competitor to step forward, win significant support from the par-liamentary party and then carry a majority of conference delegates simplyto spark a contest. Because the party conference is held at a fixed date andcannot be brought forward for a leadership challenge, the leader is vulnera-ble only once per year. Thus when a group of MPs, in January 2010, calledfor a vote of their colleagues to determine whether there was continuedconfidence in Brown’s leadership this was little more than a publicity stuntaimed at embarrassing the leader as there is no provision in party rules forsuch a vote and if one was held it would have no authority.

The Canadian parties provide many similar examples. Liberal PrimeMinister Jean Chrétien was under sustained attack within his parliamentaryparty from supporters of leadership rival Paul Martin from 2000 onwards.Chrétien could only be removed by a vote at a party convention held onceper election cycle. While it was apparent that Chrétien would likely losesuch a vote, he used his authority as leader to delay the party gathering aslong as possible. When he could put off the process no longer he announced,in autumn 2002, his planned retirement (for some 17 months later) thus cir-cumventing any removal vote. Martin’s supporters generally opposedChrétien’s remaining as party leader for this extended period but their only


options were to accept this proposition or fight a lengthy leadership removalcontest that would divide the party and if they won remove Chrétien perhapssix months earlier. Had the parliamentary party held the removal authority itis likely that this rivalry would have been resolved as much as three yearsearlier. Interestingly, after Martin became leader and Prime Minister, the partyamended its rules to eliminate any chance of a removal vote while its leader isserving as Prime Minister.

We can contrast these cases with those in which the parliamentary partyholds the removal authority. While there are many examples, we focus on the case of New Zealand Prime Minister, and National party leader, Jim Bolger as he was similarly situated to Canada’s Chrétien and provides auseful comparison. Bolger had won three elections on the trot when he waschallenged for the party leadership by fellow parliamentarian Jenny Shipley.Bolger first learned of the developing challenge upon his return from aCommonwealth heads of state meeting on a Saturday. Shipley met with Bolgerthe next day, claimed to have the support of a majority of her colleaguesand promised to force a leadership vote when the parliamentary caucus nextmet early that week. Bolger accepted defeat and made a deal that very daywith Shipley that he would immediately announce his resignation as partyleader to take effect some weeks later after attending a planned APEC con-ference (Bolger, 1998). Had the National party adopted rules similar to thosein place in the Canadian Liberal or UK Labour parties this contest wouldalmost certainly have played out in public over many months as preparationswere made for a gathering of the extra parliamentary party to resolve it.

The willingness of parties in which the parliamentary caucus holds theremoval authority to cycle through leaders is also evident in the recent case of the Australian Liberals. In the two years following their defeat in the 2007 election the party had three leaders. Outgoing Prime Minister Howardannounced his resignation on election night. Within a week he was replacedby Brendan Nelson who lasted in the leadership for ten months before beingousted and replaced by Malcolm Turnbull who in turn was ousted some 15 months later in favour of Tony Abbott. Similar if, somewhat slower,cycling through leaders can be found in other parties in which parliament-arians control leadership politics. For example, the NZ National Party, afterlosing government in 1999, churned through leaders Jenny Shipley, BillEnglish and Don Brash relatively quickly before settling on John Key whoseems to have gained some security in his position by winning the 2008 election.

We find that leaders who are either formally removed or resign underpressure are likely to have led their party to a disappointing electoralshowing in the most recent campaign. This is consistent with the literaturesuggesting that parties’ typical responses to an electoral setback includesome form of internal change (see, for example, Harmel and Janda, 1994;Harmel, 2002). Of the 17 leaders who have been formally removed, 15 led


their party in the prior election. In nine of these cases the party won alower seat share in the most recent election compared with the prior oneand in four cases won a similar seat total (within one percentage point). Asshown in Table 8.4, in only two of these 15 cases of removal had the partyperformed better electorally under the incumbent – Michael Moore in theNew Zealand Labour Party and John Bruton in Ireland’s Fine Gael. In Bruton’scase, while the party won more seats in his last campaign the election resultedin his party’s removal from government, so only in the case of Moore can itbe said that the incumbent’s last campaign was arguably a success.

We have 50 cases in which a leader resigning under pressure had led theparty in the prior election and the party had contested at least two elec-tions.6 In 32 of these cases the leader ‘forced’ from office led the party to aworse result in the most recent election compared with 13 cases in whichthe party increased its vote share and five where it remained essentiallysteady. Many of the 13 cases were also electoral disappointments despitethe parties’ increased seat share. This grouping includes John Turner(Canadian Liberals), Gough Whitlam (Australian Labor), Andrew Peacock(Australian Liberals), Neil Kinnock (UK Labour), and Michael Howard (UKConservatives). While all of these leaders saw their party do slightly better, interms of seats won in the election prior to their vacating the leadership, theresults were widely viewed with disappointment within the party. Leaderswho resigned under pressure despite leading their party to a better electoralresult also include Don Brash (NZ National) and Charles Kennedy (UK LiberalDemocrats) both of whom resigned in the face of personal scandals unrelatedto their party’s electoral fortunes. In short, leaders are often removed, or per-suaded to leave, after leading their party to a disappointing electoral outcome.

Given the key importance of who holds the removal authority to the pol-itics of deselection, it is necessary to consider the relationship betweenleadership selection and removal rules. While there is not a perfect correla-tion, as suggested above, parties that have broadened the selectorate aresignificantly more likely to have also stripped the parliamentary power ofthe sole authority to remove the leader. It is thus important to understandthe parties’ motivation in broadening their leadership selectorate, whichhas always predated expansion of the removal authority.

This is a large question that we deal with in detail elsewhere (Cross andBlais, forthcoming) and can only address in summary form here. Generally,we find that parties have expanded authority over leadership politics toinclude the extra parliamentary party when three conditions exist: they havesuffered a significant electoral defeat, they are in opposition, and a precedentexists within their party system for such an expansion. We also find that newparties are more likely to include their grassroots membership in leadershippolitics.

All of the parties that now include party members from outside the electedcaucus in their leadership politics moved to a broader selectorate while in


opposition, with the only exception being the Irish Progressive Democrats.Furthermore, in all but one of these cases, the party did so either at their for-mation or after a disappointing electoral result. The question remains, how-ever, why some parties, particularly the major parties in Australia and NewZealand have not broadened the group of formal participants in leadershippolitics. When asked, the most common reason given by party elites in thesecountries is that the three-year electoral cycle does not allow the time for along, drawn out leadership selection process. They point to leadership con-tests with broader selectorates in the United Kingdom and Canada (whereparliaments can last for up to five years) taking six months or longer andargue that with elections held every two and a half to three years theysimply cannot afford to be ‘leaderless’ for this long a period. They also pointto what might be considered a negative contagion effect. In both countries,smaller parties have experimented with balloting their members on theleadership question. Elites in the major parties point with some derision tothe experiences of the Australian Democrats and New Zealand ACT. Theyargue that in both cases these processes led to the selection of electorallyweak leaders, created unnecessary dissension within the party and did notresult in any favourable return from voters. Whether or not this criticism is well founded, what is important is the perception that the more inclus-ive selection rules were partially responsible for organizational chaos andelectoral setbacks.

We suggest that there is also an important institutional factor thataccounts for the different approaches taken in these countries. The electoralsystems used in both Australia and New Zealand mitigate the regional con-sequences of electoral setbacks. In quite a few cases of parties adoptingreforms including the broader membership in leadership politics in Canada,Ireland, and the UK, one of the compelling arguments put forward by pro-ponents of change was that the parliamentary caucus was largely confinedto a few regions. This was an important rationale in the case of the Cana-dian Liberals, United Kingdom Liberals and Conservatives and Irish FineGael. For example, the UK Tories had no MPs from Scotland or Wales at thetime of reform, Fine Gael had almost no representation from Dublin and theCanadian Liberal caucus was dominated by Francophone Canadians. In eachinstance, reformers argued that the parliamentary party was too narrow andgeographically unrepresentative to be charged exclusively with leadershipselection. The same phenomenon has not occurred in Australia and NewZealand. Representatives to the federal Australian Senate are elected via thesingle transferable vote within each state. This guarantees representation fromeach state within the parliamentary caucuses of the major parties. And asthe Australian parties allow senators to participate in the selection of leaders,they have not faced situations like those outlined above in the Canadian,British and Irish cases in which key geographic regions were not representedin the parliamentary caucus. Both the Liberals and Labor have had represen-


tation in the party room from all states for leadership elections in recentdecades. Similarly, the use of closed national lists in the New Zealand formof MMP allows the major parties to ensure representation from all parts ofthe country in their parliamentary caucuses. For example, both Labour andthe National party have had representation from the North Island, theSouth Island and the city of Auckland in every recent parliament.

Party leader accountability in an era of personalized politics:Conclusions

There is no denying the centrality of party leaders in contemporary West-minster democracies (see for example, O’Leary, 1991; McLeay, 1995; Savoie,1999). While leaders may once have been ‘first among equals’, their increasedprominence has correctly led some observers to suggest they are becomingmore akin to ‘presidents’ within a parliamentary system. In this chapter we have considered the rules and norms surrounding the removal of partyleaders in Westminster democracies. While the ability to remove a leader isnow nearly universal, we find little consistency in the methods used to holdleaders to account. The absence of state regulation of leadership politics (incontrast with some presidential systems such as the United States) providesparties with complete discretion in determining how to select and removetheir leaders. In this sense the procedures they adopt give us a glimpse intotheir democratic ethos. And the adoption of different practices results in awide variance in the security of leaders’ tenure.

We suggest that the key variables are whether some party body has theauthority to remove the leader, which group in the party has this removalprerogative and whether it is in some way checked in using it. We arguethat leaders whose parliamentary colleagues are able to depose them are farmore vulnerable than are those whose removal requires action by the extraparliamentary party. Because most leaders leave office only when facingsignificant internal pressure to do so, leadership transitions take on a verydifferent character in parties in which the parliamentary group is able toquickly depose a leader than in those in which embattled leaders are ableto hold on for lengthy periods of time because their removal requires sup-port from, and organization of, the extra parliamentary party. Thus leadersare removed both considerably more quickly and more often in Australiaand New Zealand where the parliamentary parties continue to dominateleadership politics.

Leaders respond to their institutional and political contexts in attempt-ing to increase their security in office. Accordingly, we suspect that therules determining their relative security influence the manner in whichthey interact with their parliamentary and party colleagues. Leaders whoare removable in the party room have greater incentive to continuouslyconcern themselves with maintaining the support of their fellow legislators


while those who are formally accountable to the broader party should bemotivated to cultivate support at the grassroots level. While we do notexamine the implications of different deselection rules on the behaviour ofleaders in this chapter doing so is an obvious next step as one might expectthat leaders removable by their parliamentary caucus will be more involvedin issues such as candidate selection.

The increased importance of leaders in the governing sphere coupled withthe personalization of campaigns centring around the leader suggest that aparty’s rank-and-file members should increasingly desire to be involved inleadership politics. While the general trend is in this direction, there has beenconsiderable resistance particularly in the case of leadership removal. Parlia-mentary parties who work in close proximity with leaders and whose politicallivelihood is interwoven with the success of the leader are in a unique pos-ition to judge his/her suitability to remain in the job. Party caucuses also areable to act quickly and often quietly in removing a leader preventing the longdrawn out, highly divisive affairs that often occur when these battles are takento the extra parliamentary party. Thus, there are competing interests at stakein rule-making over leadership politics. Expansion of authority to include thebroader membership may seem more democratic but it may come at real costsin terms of efficiency and party cohesion.

Notes

*We are indebted to many colleagues who have helped with this research. Among those who have been generous with their time and expertise are Raymond Miller andElizabeth McLeay in New Zealand, Paul ‘t Hart and Ian McAllister in Australia, MichaelGallagher and Michael Marsh in Ireland, and Thomas Quinn and Meg Russell in theUnited Kingdom. We have also benefited from research assistance provided by JohnCrysler, Christel Hyshka, Patrick Lemieux, Martin Croteau and Joanna Sweet.

1 Regionally-based parties, such as the Scottish National Party, whose leader’s primarylegislative position is in a sub national legislature, are excluded.

2 Interviews were conducted in Auckland and Wellington in November 2007, inCanberra in December 2007, in London in March 2008, and in Dublin in October2008. The Canadian interviews were conducted in Ottawa, Toronto and Calgaryover a longer period of time.

3 Heath and Thatcher both formally resigned after failing to secure the necessarynumber of votes to retain the leadership in the first ballot of a challenge. These arecounted as ‘removals’ as essentially the Conservative party process used at the timewas akin to the first ballot being a ‘spill’ motion which if passed (defined as theincumbent not reaching the required threshold) was followed by an open contest.This interpretation is buttressed by the allowing of new candidates to join the raceon the second ballot. In essence, both Heath and Thatcher were removed on the firstballot and open contests ensued in which they both declined to participate.

4 The Progressive Democrats ceased operations in late 2009.5 After John Bruton was removed from the leadership, Fine Gael adopted a rule

requiring a six month grace period after a leader survives a challenge. Bruton had


survived a party room vote just two months prior to the vote which succeeded inremoving him from office.

6 This eliminates Canadian Alliance leader Stockwell Day whose party ran in only onegeneral election and Australian leaders Alexander Downer (Liberal) and Simon Crean(Labor), Geoffrey Palmer from New Zealand Labour, Michel Gauthier from the BlocQuebecois, and Menzies Campbell from the UK Liberal Democrats who never ledtheir party in a general election.

Bibliography

Alderman, K. (1999) ‘Revision of leadership selection procedures in the Conservativeparty’, Parliamentary Affairs, 52(2), 260–74.

Bolger, J. (1998) A View From the Top: My Seven Years as Prime Minister (Auckland:Viking Press).

Bynander, F. and ’t Hart, P. (2007) ‘The politics of party leader survival and succes-sion: Australia in comparative perspective’, Australian Journal of Political Science,42(1), 47–72.

Carty, R. K. and Blake, D. (1999) ‘The adoption of membership votes for choosingparty leaders: The experience of Canadian parties’, Party Politics, 5(2), 211–24.

Courtney, J. (1995) Do Conventions Matter? (Montreal: McGill-Queen’s UniversityPress).

Cross, W. (1996) ‘Direct election of provincial party leaders in Canada, 1985–1995:The end of the leadership convention’, Canadian Journal of Political Science, 29(2),295–315.

Cross, W. and Blais, A. (forthcoming) ‘Who selects the party leader?’, Party Politics.Engelman, F. C. and Schwartz, M. A. (1975) Canadian Political Parties: Origin,

Character and Impact (Scarborough: Prentice Hall).Harmel, R. (2002) ‘Party organizational change: Competing explanations?’, in

R. Luther and F. Muller-Rommel (eds) Political Parties in the New Europe: Politicaland Analytical Challenges (Oxford: Oxford University Press), pp. 119–42.

Harmel, R. and Janda, K. (1994) ‘An integrated theory of party goals and partychange’, Journal of Theoretical Politics, 6(3), 259–87.

Heidar, K. and Koole, R. (2000) ‘Parliamentary party groups compared’, in K. Heidarand R. Koole (eds) Parliamentary Party Groups in European Democracies: PoliticalParties Behind Closed Doors (London: Routledge), pp. 248–70.

Henderson, J. (1981) Rowling: The Man and the Myth (Auckland: Australia and New Zealand Book Company).

Kenig, O. (2009) ‘Democratization of party leadership selection: Do wider selec-torates produce more competitive contests?’, Electoral Studies, 28(2), 240–7.

LeDuc, L. (2001) ‘Democratizing party leadership selection’, Party Politics, 7(3), 323–41.McAllister, I. (1996) ‘Leaders’, in L. LeDuc, R. G. Niemi and P. Norris (eds) Comparing

Democracies: Elections and Voting in Global Perspective (Thousand Oaks: Sage).McLeay, E. (1995) The Cabinet and Political Power in New Zealand (Auckland: Oxford

University Press).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 inWestern Democracies (London: Sage Publications), pp. 1–22.

Massicotte, L. and Blais, A. (2000) ‘Term of office, length of’, in R. Rose (ed.) Inter-national Encyclopedia of Elections (Washington: Congressional Quarterly Press),pp. 308–10.


O’Leary, B. (1991) ‘A Taoiseach: The Irish Prime Minister’, in G. W. Jones (ed.) WestEuropean Prime Ministers (London: Frank Cass), pp. 133–62.

Poguntke, T. and Webb, P. (eds) (2005) The Presidentialization of Politics: A Compar-ative Study of Modern Democracies (New York: Oxford University Press).

Powell, G. B. (1982) Contemporary Democracies: Participation, Stability and Violence(Cambridge: Harvard University Press).

Quinn, T. (2005) ‘Leasehold or freehold? Leader eviction rules in the British Conservativeand Labour parties’, Political Studies, 53(4), 793–815.

Savoie, D. (1999) Governing from the Centre: The Concentration of Power in CanadianPolitics (Toronto: University of Toronto Press).

Weller, P. (1983) ‘The vulnerability of prime ministers: A comparative perspective’,Parliamentary Affairs, 36(1), 96–117.

Weller, P. (1985) First Among Equals: Prime Ministers in Westminster Systems (Sydney:Allen & Unwin).

Weller, P. (1994) ‘Party rules and the dismissal of prime ministers: Comparative per-spectives from Britain, Canada and Australia’, Parliamentary Affairs, 47(1), 133–43.


9The Firing Line: When and Why DoPrime Ministers Fire MinisterialColleagues?Keith Dowding and Elizabeth McLeay

Moving ministers on: UK and New Zealand experiences

Prime Ministers (PMs) in parliamentary systems shape ministerial succes-sion through hiring and firing their ministerial colleagues.1 This chapteridentifies the different ways in which PMs in New Zealand (NZ) and theUnited Kingdom (UK) punish ministers who perform poorly, examiningthe similarities and differences that exist both institutionally and in thepersonality or style of the PMs involved. We explore this issue by thinkingabout the relationship between PMs and ministers in agency terms. Cabinetgovernment can be thought of as a system whereby the government isaccountable to parliament and through parliament to the electorate. Inpractice that line of accountability works through the party (Brennan andHamlin, 1993; Strøm, 2000). The PM is supported by her party (or parties incoalition governments) whilst the party gains enough public support.2 ThePM’s role is to construct and direct government on behalf of her party, andeach minister is directly an agent of the PM and through her indirectly anagent of their party.

NZ and the UK provide useful case studies through which to explore min-isterial dismissals because they share the Westminster heritage and havebroadly similar parliamentary and cabinet systems (Rhodes and Weller,2005; Rhodes et al., 2009). Both PMs run cabinet government with a dom-inant PM, supported by parliamentary majorities, and with professional‘neutral’ bureaucracies. The PMs face similar sorts of constraints when theyhire and fire. One constraint consists of the ministerial team or front benchline-up they inherit when they take over the prime ministership – their‘legacy’ cabinets. Another concerns the requirements of representative crit-eria such as geographical distribution, gender, ethnicity and, of course, ideo-logical disposition. NZ and UK PMs share much in common.

There are also important institutional differences between NZ and the UKthat also structure prime ministerial behaviour in forming and sustaining

157

their governments. Firstly, PMs are less secure in NZ than the UK. Whilst the process of challenging PMs has changed in both the major parties in theUK since World War II, their position has always been more secure than inNZ where a challenge from within the parliamentary party can occur withlittle notice. Between 1949 and 2008, ten men and two women served asPMs in New Zealand. Some had very brief tenures: Walter Nash, 1957–60;John Marshall, 1972; Norman Kirk, 1972–74; Wallace (Bill) Rowling, 1974–75;Geoffrey Palmer, 1989–90; Mike Moore, 1990; and Jenny Shipley, 1997–98.Of the 12 PMs, the National Party’s Sidney Holland and Jim Bolger andLabour’s David Lange and Geoffrey Palmer all lost their positions while stillin office, either by choosing to resign before they were pushed or through acaucus vote. Rejection by one’s peers is clearly not just a theoretical possibilitywhen parliamentary parties hire and fire their leaders.

Secondly, in NZ since the 1993–96 parliamentary term, governmentshave been, if not in formal coalitions, tacitly so, meaning the PM has totake into account other strategic factors in ministerial choice (Dowding andDumont, 2009).3 In the UK, PMs head single-party majority governments,while in NZ, since the country adopted proportional representation, PMshave had to manage more complex governing arrangements.4 Thirdly, evenin majority-party governments the NZ PMs do not always select the cabinetand are led by the party caucus to a much greater extent than has ever beenthe case in the UK.5 Fourthly, whilst the full cabinets are of roughly equalsize there are more members of the government in the UK (60–100 in theperiod we cover) than in NZ (16–28). However, due to the smaller size of the NZ legislature the ratio of ministers to ministrables (potential ministers) ismuch lower. These factors lead to a much lower turnover in NZ than the UK.Finally, turnover is affected by NZ’s shorter length of term. The UK has a tra-dition of a reshuffle mid-term, where the PM moves ministers around andbrings in new faces to renew government, a process less possible with NZ’sthree-year terms, as opposed to the UK’s up to five years. We mention theseconstraints where important but compare the selection and dismissal prob-lems of PMs in the UK and NZ in agency terms.

An agency theory perspective

According to agency theory, delegation occurs when agents have time andinformation or skills the principal lacks (Kiewiet and McCubbins, 1991;Aghion and Tirole, 1997). A principal wants her agent to do the job as ifmotivated by the principal’s preferences. We identify two agency problems:agency rent and adverse selection.

Agency rent occurs when a principal hires an agent to carry out certainactivities on her behalf, but he does not carry out those activities efficientlyor effectively. The problem can emerge because of asymmetric informationor hidden action. The agent can observe more closely what he is doing

158 The Firing Line: When and Why Do Prime Ministers Fire Ministerial Colleagues?

than can his principal. He may not work efficiently. He might not put inthe effort or lack the necessary skills. Secondly, where the agent does workhard with the required skills, he might not share the principal’s prefer-ences. In our case, the minister has different policy preferences from thePM. Following Bressler-Gonen and Dowding (2009), we call the first problem‘agent-shirking’ (or ‘shirking’) and the second ‘policy-shifting’ (or ‘shifting’).

Adverse selection occurs where there is heterogeneity in the population ofqualified candidates and those with some undesirable characteristics aremost likely to be those who come forward. In the ministerial market themost ambitious are also likely to be those who want power and/or havestrong policy preferences. The power-hungry might appeal to a PM becausethey show great loyalty in order to be chosen, but might be inefficient min-isters in other regards; or once chosen pursue their own preferred policiesrather than those of the PM. In other words, compared to less ambitiouspeople who have not caught the eye of the PM or chief whip they might bemore likely to shirk or to policy-shift once in a position of power.

Ministrables who are efficient and who also have strong policy prefer-ences are likely to gain their own following within the party. These peoplewho have to be chosen are sometimes called ‘big beasts’ (King, 1994). How-ever, if they have strong policy preferences they are also more likely to policy-shift and perhaps challenge the PM in the future. Thus adverse selectioncan be a great problem for the PM. The problem with big beasts need notbe due to asymmetric information or hidden action. To be sure, when a big beast moves against a PM he will hide his actions at the early stages, but the PM might well have known from the appointment stage that theminister might cause her policy problems. She is not fully free to choosethe cabinet of her desires because she too is an agent of her party.

How PMs deal with shirkers and shifters varies. We can describe fourbasic types: butchers, dominators, operators and consensus seekers. PMsmay exhibit more than one of these characteristics; they are non-exclusive.They emerge not only through the natural personalities of PMs but alsothrough the environment in which a PM finds herself (Dowding, 2008).These four prime ministerial styles help us to understand how PMs con-struct, manage and reconstruct their cabinets, as we explain below. Evenestablished PMs might be forced to choose ministers whom they dare notleave out of their cabinet. These big beasts are the cases where adverse selec-tion and agency rent can be easily distinguished. King recognizes, of course,that PMs dominate, and can often determine policy initiatives in areas of theirinterest. PMs often appoint ministers to specific departments in order to pur-sue policies the PM wants. Nevertheless, King argues that some ministers arerelatively autonomous and are sometimes given that autonomy by the PM.

Following King (1994), we argue that all PMs have to deal with big beasts, aparticular problem for adverse selection. Big beasts stand out because they areaccorded special status by their colleagues in cabinet and treated cautiously by

Keith Dowding and Elizabeth McLeay 159

their PMs. All have some following in their parties, and they are the ministersmost recognized by the general public. More importantly, however, each isleft largely free to run their departments, and the PM needs to get them onboard for major policy initiatives.

Firing ministers in the United Kingdom

Dealing with agency rent

Most dismissals in government occur as a result of problems of agency rent.The minister has not been up to the job. Some of these we might call ‘forcedexits’. They are the dismissals that occur following a specific identifiableproblem that is laid at the door of the minister himself. From 1945 to 2007(Attlee to end of Blair) there were 53 such forced exits. Many more exits thatoccurred following a major reshuffle can be categorized as ministers losingtheir job because of past performance. Berlinski et al. (2010) show that min-isters in the UK who have had calls for their resignation are less durable thanministers who are ‘clean’. In other words, such ministers are much more likelyto be removed during a cabinet reshuffle. This is the operation of PMs remov-ing ministers because of agency rent problems. These are resignations due toagent shirking as we have defined it.

It is much more difficult to give precise figures on removals due to policyshifting. Indridason and Kam (2008) claim that most reshuffles in cabinetoccur in order to stop such policy shifting. On their argument PMs moveministers around in order to stop them developing long-term policieswithin departments. We can see where policy disagreement leads to min-isterial exit. During the period 1945–2007 there are 68 resignations due to policy disagreement or personality clash. The former (64) are ministerschoosing to resign because they do not agree with government policy.

Some PMs such as Attlee (1945–51) were known as effective ‘butchers’ – loyal to ministers doing their job but not hesitating to get rid of thoseseen as ineffective (Thomas, 1998: 7–8). Macmillan (1957–63) was alsothought to be loyal until the ‘night of the long knives’ when he sackedeight cabinet ministers and nine junior ministers to reshape his govern-ment because of falling popularity. The move failed: the butchery was per-ceived as too bloody and done cynically for electoral reasons rather thanreasons of good government.

Margaret Thatcher (1979–90) was also seen as an effective butcher. With-out operating an inner cabinet she effectively used the cabinet committeesystem to ensure most issues were not discussed at full cabinet. She oftenspoke to ministers on a one-to-one basis and was famous for her lack of patience if they or civil servants did not quickly get to the point. Shefavoured a few ministers for advice, largely those who shared her policyaims, though William Whitelaw of somewhat different character was a ‘rock’throughout her first government. Thomas (1998: 37) says ‘Margaret Thatcher


did not leave ministers secure in their posts. By the end of her term in officeshe was the only remaining member of the original 1979 Cabinet’. Thechanges she made were not primarily on the grounds of competence but toshift the ideological balance within the government. Her stamp upon everyarea of government policy was immense. Heseltine, who later famouslyresigned, was always more of a social conservative than Thatcher, but wasseen as ‘one of us’, because of his ‘can do’ dynamism, whereas, for exam-ple, Mark Carlisle was sacked as Education Secretary despite being regardedas an effective administrator largely because he was considered to have‘gone native’. In his case policy shifting was punished.

Earlier Eden (1955–57) had dominated policy-making in his government,though less effectively, often getting bogged down in policy detail ratherthan allowing his ministers to get on with their jobs. Like Thatcher he oftensaw ministers alone, but was considered a good chair of the cabinet. Of coursehis premiership is marked by the Suez crisis, which broke him.

Wilson (1964–70; 1974–76) was an operator. He set up ministers withconvergent responsibilities so one would check the other, and liked to reshufflehis cabinet, particularly in his first administration, to ensure that no ministercould dominate policy in any one area. In his second administration heallowed ministers more discretion. Heath (1970–74) also led from the front,but his government was knocked off course by events and then largelyresponded to those events.

Major (1990–97) had problems in his second administration (from 1992),which was characterized by constant crisis. Major’s rise was rapid, and hisfall long as he presided over a deeply split party and was seen as a weak PM,perhaps unfairly given the deep divisions in his party. Much more collegialthan Thatcher, he was dependent upon other cabinet heavyweights such asHeseltine and Clarke, though relations with Clarke cooled as Major movedto the right to hold his party together. Seeking consensus he was accused ofbeing a poor butcher, not sacking ministers when he should – for exampleNorman Lamont after the fiasco of leaving ERM – and of letting personalfriendships get in the way – for example hanging on to the scandal-miredDavid Mellor. He was loath to sack ministers unless absolutely necessaryyet shed them at an alarming rate. He was criticized for dithering, thoughon becoming PM he quickly disposed of the poll tax and introduced thecouncil tax. In autumn 1994 eight ‘Eurosceptical’ Conservative MPs hadthe whip removed, but were soon brought back: Major needed their votesas his majority was disappearing. Major’s public indecisiveness was aptlysummarized in Lamont’s resignation speech: ‘We give the impression ofbeing in office, but not in power’. Major himself said that his premiershipwas ‘too conservative, too conventional. Too safe, too often. Too reactive.Later too often on the backfoot’ (Major, 1999: xxi).

Like Churchill and Wilson before him, Blair (1997–2007) was accused of cronyism. These accusations emerged early in his premiership as he


ennobled close friends and confidants. He felt the need, after 18 years ofConservative rule, to find people outside the Commons who could do gov-ernment jobs. Blair did not create more Lords than previous PMs. Later,accusations of peerages in return for donations to the party and blatantspin, notably over the evidence of Saddam Hussein and weapons of massdestruction, ruined his ‘pretty straight sort of guy’ image (self-proclaimedwhen it emerged Bernie Ecclestone had donated £1 million at the sametime Formula 1 was being exempted from the ban on tobacco sponsorship).He further downgraded the importance of the cabinet, preferring one-to-onemeetings with ministers in his famous ‘sofa government’. He made effectiveuse of cabinet committees, putting those he trusted as chairs.

Like Thatcher, Blair dominated policy in areas that caught his eye; thoughhe allowed Gordon Brown at the Treasury to take on enormous powers indomestic social policy, as the Treasury increasingly became involved inpolicy formation and implementation through its line-by-line scrutiny ofthe spending power of departments. Blair, like Thatcher, was a policy outlierin his own government, and like her suffered from a constant leakage ofministers unhappy with policy and interference. Still seen as a potentialwinner of elections however, the Labour Party seemed more reluctant to letBlair go than had the Conservatives with Thatcher. He was very loyal to hisministers, and was the only PM of the modern period to bring back a fullcabinet minister he had himself sacked. However, the conflict with GordonBrown that defined his premiership led to speculation at every cabinetreshuffle about whether he would remove Brownites and impose his ownpeople.

Dealing with adverse selection

The first cabinet of Margaret Thatcher, who by the end of her tenure wasseen as one of the most dominant of PMs, was described by some as ‘WillieWhitelaw’s cabinet’. Whitelaw had been her main rival for leadership ofthe Conservative Party. He was so loyal to Thatcher’s predecessor, EdwardHeath, that he did not stand against Heath in the leadership election, onlyentering the race following Heath’s withdrawal, by which time Thatcherhad an unassailable lead. But her first, ‘legacy’ cabinet was composed ofHeathites or, as she would later dub them, ‘wets’. However, she was able toget her supporters into key economic portfolios, and she was lucky thatWhitelaw proved as loyal to her as he had been to Heath. We can see fromthis example that PMs might have little choice over whom to have in theircabinet, but may have more freedom over where to place people – thoughnot always.

Whitelaw was more a confidant and loyal supporter of Thatcher than abig beast – though he could not have been left out and was largely auto-nomous at the Home Office. Michael Heseltine is an obvious big beast,though King suggests that Thatcher did not recognize that fact, which led to


his resignation in 1986. The other possibility is that she did recognize hisstatus and wanted him out of the cabinet as he was too big a threat to her,though his resignation seemed to surprise her at the time. Certainly, hewould have been a greater threat had he resigned over the Conservatives’disastrous poll tax in her third term.

Peter Walker is an unusual case since he was such a quiet member ofThatcher’s cabinet. He is accorded special status by King (1994: 205–6,passim) who describes Walker’s position in some detail. Walker served foralmost the entire period of Thatcher’s incumbency. A known wet anddissenter from many Thatcherite policies, he kept his head down and didthe jobs he was asked to do quietly. Without obvious reason to sack him,Thatcher kept him as he had a substantial backing in the party, representingthe Heathites she had overturned and giving them less cause to grumble.Walker was largely autonomous in both the Agriculture and Welsh Officeportfolios. Nigel Lawson might also be seen as a big beast as he was also akey figure for a while and ran the Treasury autonomously. Geoffrey Howe,originally a key Thatcher ally, became a more powerful version of PeterWalker, a quiet member of cabinet with a lot of allies in the party but not anobvious supporter of the more radical polities of his leader. Howe’s resigna-tion speech effectively ended Thatcher’s premiership.

Tony Benn is a rather different example. Continually a thorn in the sideof Wilson and Callaghan, he was a convenient bogeyman for the Conserva-tive press. However, he commanded substantial support in both the parlia-mentary and wider party. Other figures described by King are big beasts inthe sense the prime minister had no choice but to include them, some-times because they represented factions within the party (for example,James Callaghan and Roy Jenkins in Wilson’s Labour governments) or weresimply too important within the party or publicly (for example, Rab Butlerwas described as the ‘indispensable man’ in successive Conservative admin-istrations (King, 1994: 221).

Gordon Brown is one of the biggest beasts of all. He was given greatautonomy by Blair as part of the deal leading to Brown’s stepping aside in the party leadership election. Blair was unable to offer Brown anythingless than the Chancellor’s job, and Brown, backed by large factions withinthe Labour Party, was too important to be removed or shuffled. Brown used this power within the Treasury to develop domestic welfare policies well beyond the normal Treasury brief. Sometimes Blair would discoveraspects of policy through selective leaks at the same time as the public andthen Blair’s supporters would counter-leak, as each side tried to take creditfor policy initiatives. Brown acted the classic double agent (Andeweg,2000); an agent of the PM and government and an agent for his own ambi-tions. From Blair’s point of view the problem of Brown was indeed one ofadverse selection; his problem was that he had no choice but to select as he did.


In all these examples, and in many others, PMs are forced to choosecertain people as ministers because they are too powerful to be left out.They lead important factions or are supported by too many backbench MPsof the governing party to be ignored. A PM who is relatively weak, such asThatcher when she first came to power, might have her cabinet virtuallyimposed on her by the needs of the party. In John Major’s case, he wasforced to choose both Europhiles (those who wanted closer links to theEuropean Union) and Eurosceptics (those who wanted weaker links) as thatissue split the Conservatives down the middle. Having to place big beasts ina cabinet is an adverse selection problem. Big beasts are likely to be ambi-tious for the top job and are likely to have policy preferences of their own;but despite these negative features the PM cannot remove them. She is an agent of her party just as those ministers, through her, are agents of theparty.

Firing ministers in New Zealand

Dealing with agency rent

Although many NZ ministers were notable policy shirkers or shifters,remarkably few were fired. From 1949 until 2008 (covering the PMs SidneyHolland until the end of Helen Clark) there were 17 mid-term forced exits(including one minister who was sacked twice). Most occurred when PMswere no longer leading single-party majority governments (McLeay, 2009).Sir Apirana Ngata resigned in 1934 after a Committee of Inquiry criticizedhis administration of the Native Affairs Department, but no other ministerwas fired until 1982.

Apart from the scarcity of firings and the long period with none, therewas a further oddity. PMs publicly reprimanded several shirkers by remov-ing the relevant portfolios but permitted them to remain in the ministry.Also, five Labour shirkers lost their jobs at one time only to be sub-sequently reinstated, under the same PM. The small number of firings, andthe interesting risks taken by PMs in allowing shirkers to return to theministry, can partly be explained by the paucity of ministrables available, a consequence of the small, unicameral, House of Representatives (120–2MPs since 1996). Minority governments have a particularly sparse choice,especially as a minimum of three years’ parliamentary experience is theusual criterion.

With the notable exception of Roger Douglas (discussed below), policyshifters, in contrast, do not enjoy ministerial reincarnations under thesame PM, and they leave the ministry altogether rather than just lose theirportfolios: plainly, they are much greater risks for PMs than are shirkers.However, even though a number of ministers carved independent policypaths in various governments, it was not until 1982 that the then NationalParty PM, Robert Muldoon, told one of his ministers, Derek Quigley, an


outspoken critic of economic policy, to apologize or to resign. Quigley,who had been a lead player in trying to unseat Muldoon, took the lattercourse of action.

The 1984–90 Labour government saw two high-ranking ministers, RichardPrebble and Douglas, leave cabinet when they opposed the policy directionof their PM, David Lange, and criticized his leadership.6 After the caucus re-elected Douglas to cabinet when the next vacancy came up, in 1989,Lange resigned as PM, having already faced down challenges the year before(McQueen, 1991: 171–8; Rhodes et al., 2009: 130). This episode illustratesthe institutional constraints on Labour PMs who have to accept the cabinetministers elected by their parliamentary party colleagues, although PMsallocate portfolios and (since 1987) can appoint ministers (usually betweenthree and six) outside cabinet.

The next PM to dismiss a policy shifter was Jim Bolger (National Party)who sacked Winston Peters for going his own way on Maori developmentpolicy and for criticizing Bolger’s leadership. Peters then formed a new cen-trist party, NZ First. But the real problems for PMs came later, when theirparties could no longer command a majority of the House on their own.One minister left the National-NZF government (1996–98), with Bolger asPM and Peters as Deputy, because she was unhappy with its direction; anda NZF minister was sacked for opposing his senior National Party counter-part. Then Bolger was deposed by a senior minister, Jenny Shipley, who sub-sequently fired Peters for taking an opposing stance on the sale of an airportand, moreover, for criticizing her (McLeay, 1999).

Labour PM Helen Clark (1999–2008) dismissed Turiana Turia, a ministeroutside cabinet, from the Labour-Progressive minority government. Turiavoted against the controversial Seabed and Foreshore legislation and alsosaid she was resigning from the Labour party.7

The pattern by now is clear: policy shifters are no longer tolerated. Still,as the UK examples showed, being an effective butcher and a dominant PM do not necessarily go together, an interesting finding of this compar-ative study. That said, it does not always take a firing for a PM to deterpotential shifters, or bring open shifters back into line. A case in point wasRobert Muldoon who ruthlessly ruled his government between 1975 and1984. Most ministers were too cowed to oppose him. Muldoon dealt withthe shirkers not by firing them but by informing them they had no futurein the party and therefore should not seek re-election to Parliament at thenext general election. Except for the Quigley case, Muldoon avoided mid-term firing by doing his butchery in the backroom. Interestingly, Muldooncontinued the NZ tradition of making many decisions in full cabinet, andalso allowed some autonomy in portfolio management. Nevertheless, therewere constant mutterings behind the PM’s back, particularly between 1978and 1981 when there was discontent with Muldoon’s interventionist econ-omic policies (he was also Minister of Finance) and with his combative


leadership style. A plan to unseat him in late 1980 failed at the last minutewhen the only contender, Deputy PM Brian Talboys, lost his nerve (James,1986: 98–101; McLeay, 1995: 72–3; Templeton, 1995: 146–52). DespiteMuldoon’s bullying manner – or perhaps because of it – Quigley was theonly minister to be openly dismissed. Ministerial policy shifting was extra-ordinarily difficult under Muldoon, although he had his problems withbackbench dissenters.

Lange’s successor Geoffrey Palmer was one surprisingly effective butcher.Faced with inheriting Lange’s ministry – the legacy problem – Palmer adeptlymanaged a dramatic reshuffle of cabinet personnel by requiring six ministerswho had said they would retire at the next election to resign their min-isterial posts, allowing a radically different cabinet to contest the next election.This saved neither Palmer, who could not restore the public’s confidence in his government, nor his very brief successor, Moore, for Labour lost the1990 general election.

James Bolger, National Party PM, 1990–97, was both an operator and a butcher. He arrived in office with a large majority, inherited a financialcrisis, and determinedly pursued the neo-liberal economic agenda. He dis-missed Peters. But Bolger made a strategic error by promising a referendumon the electoral system. When the binding referendum was held in 1993, it confirmed the 1992 (non-binding) choice of the MMP electoral system.After the National Party only narrowly regained power in the 1993 elec-tion, Bolger moderated government policy and co-opted his most radicalministers in an effort to appease voters and to keep his government toge-ther in a parliamentary term that saw party defections and fragmentation.Bolger did not reappoint the neo-liberal Minister of Finance Ruth Richardson,to that portfolio after the 1993 election. She would accept no other (Richard-son, 1995: 178–82) and resigned before the next election. Another minister wasdropped after the election, and Bolger appointed two new cabinet ministers,bypassing the three ministers he had previously placed in the executive butoutside cabinet.

Bolger had to be adept in negotiation as well as in restructuring hiscabinet, and managed to switch from steering a radically reforming cab-inet to maintaining his party in office by manufacturing legislative coal-itions. These skills served him well after the 1996 general election and theformation of the National – NZ First government. Despite their history of antagonism, Bolger and Peters developed an easy relationship, havingnegotiated a generous split of ministerial positions for the minor party and a distribution of portfolios that put most of the spending power in the hands of the senior coalition partner. There were no signs of dissentbetween the two when various NZ First ministers showed their inexperi-ence and ineptitude and had to be dismissed. Bolger’s concessions to NZ First, however, lost him support amongst his colleagues and the primeministership.


Clark was a dominant PM, an effective operator, and a skilled butcher. Sheseldom hesitated to dismiss ministers, although she usually refrained fromacting until the evidence against them was clear. When able ministers losttheir positions, she waited until their punishment was completed, thenaccepted them back into the ministry. Apart from the dissident Turia, Clarkdismissed two ministers for mistakes related to their portfolio responsibil-ities and six for personal misconduct. Under her watch an Alliance ministerwas also dismissed. There was never any attempt to remove Clark as PM.The party perceived her electoral strength and there was no obvious con-tender. Clark was a policy PM: ministers had to learn to keep up with her.Backed by an able head of staff, Heather Simpson (they were referred to asH1 and H2) Clark quickly learned to manage minority governments.Nevertheless, her first government, a coalition with the left-wing Allianceparty, experienced difficulties when the Alliance split over personality andpolicy differences. But Clark led the government again after the 2002 and2005 elections. After 2005 she managed a minority government that wassupported by two ministers outside cabinet: Peters, NZF Leader and Ministerof Foreign Affairs, and Peter Dunne, United Future Leader and Minister of Revenue. This constitutionally unusual arrangement was remarkably successful until Peters ran into problems over his and his party’s finan-cial dealings. Clark waited months for the findings from several invest-igations into the accusations, giving rise to criticisms that she should havedismissed Peters rather than requiring him to take leave from his duties.This affair did not help the Labour-led government, and it lost the next election.

Moving away from the tougher PMs towards the more consensual ones,Sidney Holland was a dominant National Party PM (1949–57) who led hisparty to its first success after being created in 1936 in an effort to unite theanti-Labour forces. Holland called an early election in 1951. His excuse wasthe bitter waterfront dispute that he then brutally suppressed using thearmed forces and authoritarian legislation. Having won this election, he alsowon the next. Despite his outward aggression and toughness, in cabinetHolland clung to ministers loyal to him. He was not an effective butcher.When in 1954 a minister of Industries and Commerce became embroiled in a public conflict of interest, the PM merely required an exchange of therelevant portfolios. Surprisingly, one of Holland’s actions was to abolish theappointed upper house, thereby diminishing his patronage opportunitiesand the range of ministrables.

The next National Party PM, Keith Holyoake, was a consensus seeker whoheaded a cabinet full of strong individuals. He took ‘straw polls’ of his par-liamentary colleagues on the cabinet line-up (although he did not alwayscomply with the majority view) (McLeay, 1995: 70). Holyoake was loyal tohis colleagues, with sometimes counterproductive effects because alongsidethe major talents sat some very weak ministers whom he was loath to fire.


Holyoake also expected ministers to be loyal to him. He scarcely ever reshuffledthe pack (Wood, 1997: 39).

Another consensual, collegial PM was Lange who defeated Muldoon’sNational Party government at the 1984 snap general election. Lange learnedtoo late how to prune his cabinet. Clever, witty and articulate, he was gen-erally content to let his talented ministers go their own way, forging aneconomic and public sector revolution, until they went too far. Althoughhis cabinet had been elected by the caucus, he expanded his patronage by creating positions outside cabinet which he appointed himself, initiallyappointing six undersecretaries. Lange’s problems really began after thehighly successful 1987 general election. Retirements created four vacanciesbut when the time came to elect the new cabinet no sitting ministers losttheir positions. Again, though, Lange changed the rules to get the team hewanted by getting permission from caucus to appoint four ministers outsidecabinet – an innovation for NZ – and appointing three undersecretaries.

Despite expanding the number of ministers with personal loyalties to him-self, Lange had a major problem with policy shifters. Although he reshuffledportfolios to reduce the power of the radically neo-liberal economic ministers,the leaders of that group finally went too far, especially when they pushed fora flat tax. Lange dismissed the Minister of State-Owned Enterprises, RichardPrebble, who had not only promoted the sale of assets more energeticallythan the PM (and many other Labourites) wanted, but had also criticized the PM himself. The Minister of Finance, Douglas, shortly after said he couldno longer work with Lange, and Lange took this as a letter of resignation(McLeay, 1995: 66–8). A Douglas ally, ironically a minister outside cabinet,resigned from his position. As noted, after Douglas was elected back intocabinet in 1989, Lange resigned. This example is a textbook case of the prob-lems of principal-agent relationships complicated by particular institutionalrules.

Dealing with adverse selection

Adverse selection is highly likely when the pool of ministrables is small,when PMs are selected by the parliamentary colleagues and therefore must beloyal to them, and when, as with Labour, PMs must accept their colleagues’choice of cabinet ministers.

Although PMs did not fire their poor ministers until the early 1980s,most reshuffled their ministries at the relatively high rate predicted anddemonstrated by Kam and Indridason (2005). PMs such as Holyoake,Muldoon and Clark were adept at persuading ministers to retire quietly at the end of the parliamentary term. Thus the triennial term can be seento have a two-way effect on the relationship between PM and minister: it provides a formal occasion and excuse for reshuffling the pack, discard-ing unsuitable ministers and demonstrating team renewal; but it inhibitsmid-term dismissals, as opposed to retention and reshuffle. The risks to


PMs elected by parliamentary parties are high. There is little time to rebuildparty and public support after dealing with one’s adverse selections.

Reshuffles are particularly likely when PMs find themselves with legacycabinets. This might be the cabinet chosen by one’s predecessor (a price fortaking over mid-office) or the team of frontbenchers and previous ministersbuilt up while in opposition. But there’s also the situation when the legacycabinet is one’s own: the tired line-up that needs a new look. Muldoon radically switched portfolios around to get the agents he wanted. He hadtwo criteria: loyalty to himself (and he overworked some loyalists, leadingto policy-shirking) and compliance with his policy direction (Templeton,1995). Clark in 2007 had three cabinet vacancies created by an impendingretirement, a return to the backbenches, and a dismissal, and a further gapbecause a minister outside cabinet resigned his warrant. Not only were thevacancies filled but Clark also changed around portfolio responsibilities.She said, ‘More than half the Cabinet Ministers named today were not in Cabinet in December 1999, and close to half were not in the Cabinetelected after the 2002 election’ stressing the regeneration of the party at alllevels (Clark, 2007).

Representative criteria can also lead to adverse selection. For example,Holyoake had a keen although blinkered awareness of representativenesscriteria in his cabinet line-up: a rural-urban balance; a regional distribution;but a neglect of the growing demand for women’s representation despitehaving an appointable woman in caucus. Clark, too, saw representative-ness, including Maori ministers, as important (Clark, 2007). Thus PMsconfine their pools of ministrables.

NZ PMs have had their share of big beasts. Holland was lucky thatHolyoake, a potential threat, lost his seat in 1938 and was out of Parliamentfor five years. This meant that he could not contest the leadership in 1940,and Holland won (Gustafson, 1997: 132–5). According to Gustafson (1997:135), Holland would have preferred his own favourite, Bill Sullivan, to havebeen his Deputy when his de facto Deputy retired in 1946. After consultingthe members of caucus (for there was no formal election at that stage) Hol-land found that Holyoake had the numbers. The dominant faction (harken-ing back to the pre-National Party parties) was Holland’s own, however,which left Holyoake somewhat on the edge, and relationships between thePM and his Deputy were cool (Gustafson, 1997: 136). Nevertheless, whenHolland had been persuaded to retire shortly before the 1957 election,Holyoake was unanimously declared the new leader and PM (Jackson, 1978).

Holyoake himself had a cabinet full of big beasts – Ralph Hanan, TomShand, Peter Gordon, Brian Talboys, Robert Muldoon, and his Deputy, JohnMarshall –whom he managed by letting them roam, within the boundariesof collective cabinet norms and practices. But he did not hesitate to over-ride their decisions if necessary, and there were some major policy rows. Inretrospect it is astonishing that none of his ministers challenged for the


leadership, especially between 1969 and 1972. But several big beasts becameill or died before the time was right. Moreover Holyoake had retained the trustof the party, which made it difficult for others to compete against him (seeGustafson, 1997: 140), although there was an attempt to get him to resign in1971. In early 1972 Holyoake finally gave in to the pressure, and Marshalltook over. But the latter’s tenure was brief because the National Party lost theelection that year and then Muldoon successfully contested the leadership.

As noted above, for Muldoon there could be only one big beast, himself.He bullied and corralled possible challengers. Lange managed his big beastsineffectively, particularly Prebble and Douglas, although neither succeededhim. Palmer was a reluctant PM, having been content as Lange’s Deputy.Bolger’s appointment of the maverick Peters in 1990 was always a risk, but he had a high profile and Bolger wanted a Maori in cabinet. Bolger did not hesitate to get rid of him when the time came, however, as did JennyShipley after she had taken over from Bolger. Clark had some strong per-sonalities in cabinet, but none posed a risk. Her senior ministers were loyalto her personally and to the direction of policy. Michael Cullen, her veryable and articulate Deputy and Minister of Finance, was not interested in the top job, and Clark quickly rid herself of a potential big beast, JohnTamihere, by dismissing him as soon as she had good cause.

Conclusions: Prime ministers and ministerial turnover inWestminster

All PMs are vulnerable to agency rent: their agents may shirk their res-ponsibilities or shift policies away from the direction desired by their prin-cipals. Delegation is a tricky process and the consequences are not alwayseasily predicted. As these two case studies have shown, PMs in Westminstersystems have to make strategic and tactical decisions as to when, and if, tofire shirkers and shifters, and when to reshuffle portfolio responsibilities.When making these decisions, PMs must consider the seriousness of theundesired behaviour, the timing of the next election, the impact on theirown public reputation, the impact on the ministry itself, and the effects ontheir relationships with their parties to whom they are responsible.

The NZ and UK PMs discussed here have had to manage the problemsthat arise from adverse selection. None ever achieved their dream team,partly because they inherited ministers from their predecessors, and partlybecause they had to heed seniority, a criterion for appointment that mightappear to be safe enough but that can lead to poor performance. Further, as we have argued, a PM cannot always ignore the claims of the big beastswho, after appointment, might threaten governmental policy priorities andthe tenure of the PM herself. Informal rules about representative criteria(region, ethnicity, gender, ideology) must be borne in mind by PMs, despitethe possible adverse effects on ministerial selection.


Institutional factors structure prime ministerial behaviour, especiallyministerial selection and firing. In this, there are major differences betweenNZ and the UK: the vulnerability of PMs to challenges; whether or not PMslead single-party majority or coalition governments; whether PMs can picktheir own people or must accept their party’s choices; the sizes of the legis-latures, and the lengths of the parliamentary terms. Despite the differentinstitutional circumstances under which PMs in these two states con-struct, manage and reconstruct their ministries, they all have to managethe consequences of adverse selection.

Leadership duration and succession are closely bound into the relationshipsthat develop between principal and agent, particularly the ways in which PMsmanage the careers of their parliamentary colleagues: appointing ministers,distributing portfolios, reshuffling responsibilities, and firing shirkers andshifters. Institutional factors affecting hiring and firing decisions constrainPMs but each PM has his or her personality and to some extent shape theirgovernment team by their own characteristics. The UK and NZ PMs discussedhere exhibit distinctive styles in the way in which they perform these tasks,styles that interestingly cut across the institutional and contextual differences,as we demonstrate with our sketches of PMs in both countries.

We identified four non-exclusive styles of PMs – butchers, dominators,operators, and consensus seekers – to try to bring some order to our dis-cussion. PMs combine these leadership styles to deal with their ministerialshirkers and policy shifters, to manage and control their ambitious col-leagues and to refresh their governments. There is no ideal leadership style.PMs need to adapt to the circumstances in which they find themselves – the cabinet they (partly) inherit, the economic and social circumstancesof their era and importantly the problems that arise through the behaviourof their cabinet colleagues. How PMs exercise these judgements and howskilfully they use their style vis-à-vis their cabinet colleagues affects the fatesof their governments as well as their own political futures.

Notes

1 We would like to thank participants at the seminar on chapters for their commentsespecially Marian Simms, John Uhr and Paul ’t Hart. We also thank Anne Gelling.

2 In keeping with usual practice in agency theory we refer to the principal as a ‘she’and the agent as a ‘he’.

3 The first election held under proportional representation rules was in 1996, butthe previous parliamentary term saw many party defections that impacted ongovernment management.

4 Whilst in production a coalition government between the Conservatives andLiberals formed following the May 2010 UK general election. We do not analysethat government in this chapter.

5 UK Labour shadow cabinets have been elected and when taking power PMs havebeen expected to form their cabinet from the shadow cabinet, but they have movedministers around or strategically shifted powers across departments. In both the UK


and NZ factions within parties, whilst an obvious constraint, are not as important asin many other countries in terms of cabinet formation.

6 Quigley, Prebble and Douglas all became involved in the right-wing ACT partythat supported the National Party in government after the 2008 election.

7 Turia later re-entered Parliament as leader of the Maori Party and supported theNational Party after the 2008 election.

Bibliography

Aghion, P. and Tirole, J. (1997) ‘Formal and real authority in organizations’, Journalof Political Economy, 105(1), 1–29.

Andeweg, R. B. (2000) ‘Ministers as double agents? The delegation process betweencabinet and ministers’, European Journal of Political Research, 37, 377–95.

Berlinski, S., Dewan, T. and Dowding, K. (2010) ‘Performance evaluation of Britishministers’, Journal of Politics, 72(2), 1–13.

Brennan, G. and Hamlin, A. (1993) ‘Rationalizing parliamentary systems’, AustralianJournal of Political Science, 28(3), 443–57.

Bressler-Gonen, R. and Dowding, K. (2009) ‘Shifting and shirking: Political appoint-ments for contracting out services in Israeli local government’, Urban AffairsReview, 44(6), 807–31.

Clark, H. (2007) ‘Renewed cabinet line-up’, Press statement, 31 October. Accessed athttp://www.beehive.govt.nz/node/31156 Downloaded 26/10/2009.

Dowding, K. (2008) ‘Perceptions of leadership’, in P. ’t Hart and J. Uhr (eds) PublicLeadership: Perspectives and Practice (Canberra: ANU E Press), pp. 93–102.

Dowding, K. and Dumont, P. (2009) ‘Structural and strategic factors affecting thehiring and firing of ministers’, in K. Dowding and P. Dumont (eds) The Selection ofMinisters in Europe: Hiring and Firing (London: Routledge), pp. 1–20.

Gustafson, B. (1997) ‘Holyoake and the National Party’, in M. Clark (ed.) Sir KeithHolyoake: Towards a Political Biography (Palmerston North: The Dunmore Press).

Indridason, I. H. and Kam, C. (2008) ‘Cabinet shuffles and ministerial drift’, BritishJournal of Political Science, 38(4), 621–56.

Jackson, K. (1978) ‘Political leadership and succession in the New Zealand NationalParty’, in S. Levine (ed.) Politics in New Zealand (Sydney: Allen & Unwin),pp. 161–81.

James, C. (1986) The Quiet Revolution (Wellington: Allen & Unwin/Port Nicholson Press).Kam, C. and Indridason, I. H. (2005) ‘The timing of cabinet reshuffles in five parlia-

mentary systems’, Journal of Legislative Studies, 30(3), 327–63.Kiewiet, R. D. and McCubbins, M. (1991) The Logic of Delegation (Chicago: University

of Chicago Press).King, A. (1994) ‘Ministerial autonomy in Britain’, in M. Laver and K. Shepsle (eds)

Cabinet Ministers and Parliamentary Government (Cambridge: Cambridge UniversityPress), pp. 203–25.

Major, J. (1999) The Autobiography (London: Harper Collins).McLeay, E. (1995) The Cabinet and Political Power in New Zealand (Auckland: Auckland

University Press).McLeay, E. (1999) ‘What is the constitutional status of the New Zealand cabinet

office manual?’, Public Law Review, 10(1), 11–17.McLeay, E. (2009) ‘Regime change and rule adaptation: Prime Ministers and ministerial

dismissals in New Zealand’. International Political Science Association Congress,Santiago, Chile, 12–16 July.


McQueen, H. (1991) The Ninth Floor: Inside the Prime Minister’s Office – A PoliticalExperience (Auckland: Auckland University Press).

Rhodes, R. and Weller, P. (2005) ‘Westminster transplanted and Westminster implanted:Exploring political change’, in H. Patapan, J. Wanna and P. Weller (eds) WestminsterLegacies: Democracy and Responsible Government in Asia and the Pacific (Sydney:University of South Wales Press), pp. 1–12.

Rhodes, R., Wanna, J. and Weller, P. (2009) Comparing Westminster (Oxford: OxfordUniversity Press).

Richardson, R. (1995) Making a Difference (Christchurch: Shoal Bay Press).Strøm, K. (2000) ‘Delegation and accountability in parliamentary democracies’,

European Journal of Political Research, 37(3), 261–89.Templeton, H. (1995) All Honourable Men: Inside the Muldoon Cabinet 1975–1994

(Auckland: Auckland University Press).Thomas, G. P. (1998) Prime Minister and Cabinet Today (Manchester: Manchester

University Press).Wood, A. (1997) ‘Holyoake and the Holyoake years’, in M. Clark (ed.) Sir Keith Holyoake:

Towards a Political Biography (Palmerston North: The Dunmore Press).


10Power Consolidation in LeadershipChange Contexts: A Social IdentityPerspectiveEmina Subasic and Katherine J. Reynolds

Mother: What’s the matter Zoe? Why are you crying? Child: [Zoe, 4 years old, tears streaming down her face] Because

I just want Tony Blair to come back and I want him to bePrime Minister.

Mother: Why? Child: Because I just do … Mother: Why do you want him to be Prime Minister still? Child: Because I love him. [Sobs.]Mother: But there’s gonna be a new Prime Minister … There’s gonna

be a new Prime Minister, darling, called Gordon Brown. Child: I know … Mother: You don’t want him? Child: No, I want Tony Blair …

Transcript of a YouTube video titled ‘A very sad day …’, with a caption ‘A 4 yearold’s extreme reaction to Tony Blair’s departure as Prime Minister’.1

Introduction

Whether or not Zoe’s reaction epitomized the mood of the British nationon the 27 June 2007 is an empirical question.2 Even in the absence of suchextreme reactions from the general public, however, there was little doubtthat Gordon Brown’s task in succeeding Tony Blair was a difficult one. Aboutthe same time in Australia, after a decade as Treasurer and quite a few years ofharbouring hopes of becoming the next Liberal Party leader and PrimeMinister, it was becoming clear that Peter Costello’s chances of succeedingJohn Howard were rather slim. Fast-forward to early 2008 and the morning ofthe apology to Indigenous Australians, Costello may have counted himself alucky man. Had he succeeded Howard, it would have been him, rather thanBrendan Nelson, who had ‘the toughest job in politics’ – to address the nation

174

as opposition leader immediately following PM Rudd’s eloquent andmoving apology to Indigenous Australians. As Nelson spoke in defence ofHoward’s legacy and the Liberal Party’s long-standing opposition to suchan apology, thousands of spectators outside Parliament House (and aroundthe country) first booed and then turned their backs to the massive tele-vision screens broadcasting his message. In the months that followed,Brendan Nelson’s leadership became increasingly precarious – he was suc-cessfully challenged and replaced by Malcolm Turnbull in September 2008who in turn was toppled in December 2009.

What these events make clear is that leadership succession (and tran-sition) in government are often defining moments not only for leaders, butalso their political party and national constituencies. As such, contemporaryresearch in political science, political psychology and other disciplines hasexamined (albeit in a somewhat fragmented fashion) their origins, dynamicsand consequences (e.g. Giambatista et al., 2005; Bynander and ’t Hart, 2006).The goal of this volume is to integrate some of these efforts. A key feature ofdemocratic societies when it comes to political leadership is the ‘perpetualdemand to legitimise’ (Kane and Patapan, 2008: 32) and this is particularlytrue in contexts of leadership change. As such, the aim of this chapter is tobetter understand the (social psychological) processes that affect whether ornot leaders will be able to consolidate their power – as influence and (legit-imate) authority – in the context of leadership transition and/or succession.

Leadership change as a social identity process

Unlike many other political and organizational dynamics, ‘old’ leadersdeparting and ‘new’ leadership ascending to a position of power readilycapture the attention of the broader public. The recent national leadershiptransition in the United States (from Republican George W. Bush to DemocratBarack Obama) and Australia (from John Howard and the Liberal/NationalCoalition to Kevin Rudd and the Australian Labor Party) are salient exam-ples. Much of the political and social commentary surrounding these publicevents has to do with whether or not ‘new’ leaders have ‘what it takes’ fortheir prospective jobs. Features such as charisma, vision, particular personalitycharacteristics, temperament and behavioural styles are regularly evoked asarguments in support of or seeking to undermine particular candidacies. Suchindividualistic approaches to leadership feature prominently not only inpopular but also social-scientific discourses. It is worthwhile noting, though,that even these judgements about individuals (e.g. Obama as charismatic) arealways made in comparison to the available leadership alternatives, includingincumbents, as well as in light of the broader national and international,social, political and economic dynamics facing leadership candidates.

We have written elsewhere about the growing recognition that leadershipcannot be understood in isolation from the relationship between leaders and

Emina Subasic and Katherine J. Reynolds 175

followers (e.g. Turner et al., 2008). At the core of leader-follower relations isa sense of psychological connection in terms of shared social and psycho-logical group memberships, where leaders are those group members whoare seen to best embody ‘our’ shared goals, values, beliefs and aspirations(Turner and Haslam, 2001; Reicher et al., 2005). Understanding how such apsychological connection between leaders and followers is created, main-tained and lost is fundamental to understanding leadership and leadershipchange dynamics. In the remainder of this chapter, we propose a frame-work for understanding leadership transition and succession as group andintergroup phenomena. More specifically, we look at the social identityprocesses that make it possible for leaders to consolidate their power.

Social identity is that aspect of the self derived from one’s membership of social groups and the value or emotional significance attached to thosememberships (Tajfel and Turner, 1979). Rather than thinking of ourselvesin purely personal or individual terms (e.g. ‘I’, ‘me’), the self can also bedefined in terms of contextually relevant or salient (i.e. psychologicallymeaningful) group memberships (e.g. ‘us Australians’, or ‘us Liberals’). As such, self-perception can be seen as varying along an interpersonal-intergroup continuum, so that, across social contexts, people shift betweendefining themselves as individuals (i.e. in terms of their personal identity)to self-definition as group members (i.e. in terms of their social identity).

Those people with whom one shares a relevant or salient social identityare seen as ingroup members, while those excluded from group member-ship (on some relevant comparative dimension) are seen as outgroup mem-bers. For example, if the relevant comparative context concerns politicalorientation, thinking of oneself as a ‘left’ or ‘cosmopolitan’ rather than‘right’ or ‘nationalist’ voter makes sense. As such, the emergence and defin-ition of one’s social identity is a contrastive and comparative process, so that one’s ingroup tends to be defined with reference to a relevant out-group (e.g. left vs. right, men vs. women, Republicans vs. Democrats).Defining the relevant outgroup also comes to clarify and redefine who isincluded in the ingroup – and vice versa.

Furthermore, social identity or self is hierarchically organized so thatlower-level distinctions between groups (e.g. Democrat vs. Republican) becomeless important or relevant when one identifies with a relevant higher-level,superordinate identity (e.g. as Americans). Such a shift in self-definition isdirectly linked to followers’ perception of the social context – whether it isseen in terms of intergroup (e.g. factions within a political party; interpartyconflict) or intragroup (e.g. intraparty unity; national solidarity) relations.This insight from self-categorization theory (Turner, 1982, 1985; Turner et al., 1987) has been applied extensively in the area of prejudice reduction,for example, as it has been shown that encouraging re-categorization (e.g.through intergroup contact) of those initially seen to be outgroup membersas included in some relevant, higher-level ingroup, reduces inter-(sub)group

176 Power Consolidation in Leadership Change Contexts: A Social Identity Perspective

animosity and prejudice, and enhances helping and cooperative relations(Gaertner et al., 1993; Dovidio et al., 1997; Levine et al., 2005). By impli-cation, therefore, creating social division (within or between groups) – forexample, by re-defining the relevant identity so that some ingroup membersare no longer seen as part of ‘us’ – is likely to attenuate cooperation andinstead enhance conflict, animosity and prejudice. We will come back to this point in discussing how leadership processes affect whether self-definition takes place at a subgroup or superordinate level and what theimplications are for the nature of intergroup relations but also leader’scapacity to influence.

What is important to clarify now is that ingroup/outgroup and intra-group/intergroup distinctions are important to consider because we alsoknow that influence – in the form of capacity to shape other’s views andbeliefs and exercise legitimate authority – flows from shared identity anddisappears when such an identity is severed (e.g. Haslam and Platow, 2001;Turner, 2005; Subasic et al., forthcoming). So, when people think of them-selves as members of particular social groups (i.e. self-categorize), they aremore likely to be influenced by others who also share their social (psycho-logical) group membership (Turner et al., 1987; Turner, 1991). It is nowwell-documented, for example, that leaders who are seen to share the rel-evant social identity with followers (group membership, but also norms,values and beliefs that define who ‘we’ are) have greater legitimacy, sup-port and endorsement, but also capacity to influence followers’ beliefs and attitudes (Haslam and Platow, 2001; Haslam et al., 2001; Platow and Van Knippenberg, 2001; Duck and Fielding, 2003; Van Knippenberg andVan Knippenberg, 2005; Subasic et al., forthcoming).

Centrally, which particular ingroup member will have the greatest capacity to shape the group’s goals and objectives – in other words, exerciseleadership – is a function of a meta-contrastive judgement made withina particular group but with an eye to the broader context of intergrouprelations. As such, leadership is a function of being seen to best represent‘our’ goals, norms and values compared to other group members. Suchgroup goals, norms and values are emerging as important and self-defining(who ‘we’ are) given the group’s position within the broader context of inter-group relations (who ‘they’ are; the presence of external threats to the group’sstatus or wellbeing by ‘others’). The concept of prototypicality is meant to capture the extent to which a particular group member is perceived as embodying the relevant (ingroup) norms, values and beliefs given thesocial reality of group life and intergroup relations.

Leadership influence stems from a shared belief between leaders and followers that a particular action is valid, appropriate and right and is seen as the strongest form of power (see Turner, 2005). Like influence, legitimateauthority has shared social identity at its core, so that followers support aleader’s position and comply with their requests because of a shared belief


that a leader has the right (conferred by the group) to make particular deci-sions, even though one may not be fully persuaded by or agree with theirview. Finally, coercion involves forcing others to act against their will andis closest to the way in which power is defined within the ‘standard’ view.Coercion is seen as the weakest form of power as it has the capacity to severthe shared identity between leaders and followers and as such undermineinfluence (e.g. see Subasic et al., forthcoming). The social identity analysisof power is particularly relevant in explaining how influence and legitimateauthority are gained, maintained and lost – a key question when it comesto understanding leadership change.

Leadership change, social identity and power consolidationdynamics

Influence or persuasion, as well as legitimate authority, are based on the per-ception of shared identity or psychological group membership between thesource and target of power. When it comes to power consolidation, there-fore, those leadership strategies that enhance a shared identity with fol-lowers will also consolidate and enhance a leader’s power. More specifically,within the social identity perspective leadership is conceptualized as relativeinfluence and power within a group where leaders are those group membersperceived as relatively more prototypical or representative of ‘who we are’and as such more influential (Turner and Haslam, 2001; Haslam, 2004;Turner et al., 2008). People follow leaders because they embody and define‘our’ norms, values and beliefs in a particular social context, including how‘we’ should relate to ‘them’ given the reality of social relations. Thoseleaders who violate followers’ understanding of either the relevant socialidentity (i.e. ‘who we are’) or social reality are likely to engender oppositionand challenge to their position (Reicher et al., 2005).

However, rather than simply reflecting the relevant group values andbeliefs, leaders are ‘identity entrepreneurs’ who actively seek to shape thegroup identity in a way that maintains and strengthens their position(Reicher and Hopkins, 1996a, 1996b, 2001; see also Subasic et al., 2008). In this process, leaders actively construct and re-define not only who isincluded in the relevant ‘us’ (identity or category boundary) but also whatit means to be ‘us’ (identity content or meaning).

Far from being confined to the rhetorical redefinition of the relevantingroup identity, leaders can be active agents in the creation and exacer-bation of distinctions and tensions with outgroups as well as the creationof solidarity and shared purpose amongst erstwhile adversaries. Leaders attimes resort to strategies that capitalize on social division and conflict. Forexample, it could be argued that the ‘politics of division’ (e.g. ‘playing therace card’, ‘fear campaigns’) is fundamentally about creating social conflictthrough the exclusion of some relevant ‘other’. Leaders may be particularly


likely to do so when faced with threat, challenge and the possibility ofbeing ousted. Equally, however, leaders may emphasize the rhetoric of unityand solidarity with the aim of transcending existing intergroup conflict asan integral part of asserting and consolidating their leadership. Far frombeing just agents of confrontation (e.g. Rabbie and Bekkers, 1978) leaderscan be key forces in ‘bridging the divide’ and bringing about greater inter-group harmony (e.g. Pittinsky and Simon, 2007; Pittinsky, 2009). To under-stand leadership consolidation following transition or succession morefully, it is useful to consider these dynamics as interdependent and co-occurring (Subasic et al., 2009a).

From a social identity perspective, division into an ‘us’ and ‘them’ and‘their’ exclusion simultaneously strengthens the shared identity betweenleaders and the rest of ‘us’ as followers (i.e. enhances ingroup solidarity). Itthereby mobilizes ingroup support to advance a particular agenda. But howis such division achieved in the first place? Social identities emerge and aredefined through social comparison. By changing the comparative dimensionalong, for example, racial, class or gender lines the meaning and member-ships of relevant social categories are altered. As a result, erstwhile allies (e.g.Whites, middle classes, men) may redefine themselves as one another’sadversaries – and vice versa. Social comparison can also vary with changesin the relevant comparison outgroup. What ‘being (an) Australian’ is takento mean, therefore partly depends on whether one thinks of Australia inrelation to say, the US, Indonesia or France.

Modifying the comparator or the context of comparison are social creativitystrategies, traditionally recognized within social identity theory as an effectiveway for disadvantaged group members to see their group in a more positivelight by altering the elements of the relevant social comparative context(Tajfel and Turner, 1979). Social creativity – changing the comparativedimension, the values assigned to the group attributes, or the comparisonoutgroup – is also an influence process through which (emerging) leader-ship can mobilize (or de-mobilize) different groups of followers in line withparticular agendas and projects. For example, when John F. Kennedy,standing in front of a huge crowd at West-Berlin’s city hall declared that‘Ich bin ein Berliner’, he was sending a powerful message to both West andEast. Likewise, one of the key challenges for Barack Obama as both can-didate and president has been to maintain the focus on those social com-parative dimensions and definitions of who ‘we’ are that would strengthensupport for him and his policies (e.g. class, egalitarianism), while down-playing the relevance of other dimensions that could potentially divide theconstituency in ways detrimental to his leadership (e.g. race, immigrant).

Furthermore, ‘we’ or ‘us’ is always defined with reference to relevantothers who are not ‘us’ – and vice versa. The interdependence and reciprocitybetween inclusion and exclusion is rarely considered. While it is recognizedthat strategies such as ‘playing the race card’, for example, seek to create or


exacerbate social fault lines, their ultimate (and overlooked) purpose is tounite and create particular forms of solidarity at the level of the ingroup.Such solidarity, we argue, is achieved not through an ‘us’ vs. ‘them’ divideper se but a consequent and simultaneous re-definition of ingroup member-ship and the associated goals, values, norms and beliefs of what it means to be ‘us’. Equally, emphasizing core ingroup goals and values (e.g. what it means to be Australian) not only can enhance group cohesiveness andunity around who ‘we’ are and what ‘we’ do. It also clearly denotes thatthose who are out of line with ‘our’ goals and values are excluded fromgroup membership (by using a values test to determine who becomes anAustralian citizen, for example).

Strategies such as ‘divide and conquer’ have been traditionally under-stood as primarily targeting one’s political opponents, seeking to divide themon a particular issue and as such undermine their capacity to act as a unitedfront (Fear, 2007). However, leaders can also (at least attempt to) mobilizeingroup solidarity and support by creating and/or redefining the relevantingroup vs. outgroup distinctions through creation and exclusion of particularoutgroups (Subasic et al., 2009b). Ideally, such strategies will not only invigor-ate the commitment of existing supporters but also cut across ‘traditional’ distinctions to extend the support base and include in the relevant ‘us’ thosegroup members who otherwise may be reluctant to endorse the leader.

When coupled with perceptions of outgroup as threatening to ‘our’ wayof life, exclusion not only serves to clarify who ‘we’ are and what ‘we’ standfor but also positions the leader as the ‘saviour’ or defender of the group’sidentity. Prime minister John Howard’s move to send specialized militaryunits to prevent asylum seekers from reaching Australian immigration zoneshortly before the 2001 federal election is one example of an incumbentleader in self-styled ‘saviour’ mode. Many have argued that this move, madein response to a high profile incident aboard one ship carrying asylumseekers around that time played a major role in Howard’s electoral victory(cf. McAllister, 2003). It not only solidified the support of those Liberalvoters already opposed to immigration, but also others who might have sup-ported Labor in a different context but found their border protection atti-tudes (and fears, see De Castella et al., 2009) mirrored in Howard’s ‘toughstance on illegal immigrants’.

Outgroup exclusion and derogation will serve the purpose of enhancing a leader’s influence and authority only to the extent that they are seen to be in line with the shared norms, values and beliefs that define who ‘we’are and, as such, are seen to be legitimate. Therefore, paradoxically, the rhe-toric of ingroup victimization is often accompanied by a parallel emphasis on ingroup strength and virtue (Reicher et al., 2008). Glorifying the ingroupserves to justify radical ‘rescue’ strategies advocated by leaders, so that out-group marginalization, and in some cases annihilation, becomes a legitimateway of defending ingroup interests. As such, the same leadership behaviours,


decisions and policies that seem coercive and dictatorial from an ‘outsider’perspective, are likely to be seen as justified and legitimate by an ingroupaudience comprised of those who share a leader’s vision.

When outgroup exclusion is successfully legitimized as being in the bestinterests of the ingroup, it strengthens the shared social identity betweenthe leader and followers while simultaneously curtailing potential chal-lenges to leadership and, therefore, consolidates one’s power and influence.This approach is particularly useful when the excluded ‘other’ comprisesvocal opponents of the leader’s policies and practices. For example, JohnHoward’s continued emphasis on the distinction between the so-called ‘elites’and ‘ordinary Australians’ allowed him to effectively and ‘legitimately’ mar-ginalize some of the most articulate critics of his regime as ‘out of touch’ withthe proverbial ‘Aussie battler’ and as merely standing up for ‘political cor-rectness’. Importantly, such rhetoric also allowed Howard to position himselfas the embodiment of the beliefs and values, views and wishes, of the ‘silentmajority’ of ‘ordinary Australians’ (Brett, 2005).

Hence the politics of division is equally about building solidarity andcohesiveness among existing and newly recruited members of the con-stituency. As a ‘them’ is created and excluded from ‘us’, who ‘we’ are iscrystallized and solidified in line with a particular agenda. Who will bedivided and who united, who included or excluded, depends not so muchon the individual whims of leaders but rather on the relevant social context(the groups involved and the relationships between them) in which leader-ship (and its transition and succession) takes place. Leaders’ agency is con-strained by the nature of relevant intra and intergroup relations withinwhich they operate, as well as by the structural constraints and oppor-tunities provided by the institutions and offices they inhabit. We revisitthese points in more detail below.

Leadership transition and succession: A social identityperspective

As defined in the introductory chapter (’t Hart and Uhr, this volume),leadership transition in government arises out of interparty competition. At its core lies the transfer of governing power from one group of hands to another (e.g. Democrats replacing Republicans). In contrast, successionoften emerges out of intraparty competition between incumbent leadersand those aspiring to leadership. It can take place when a party is in as wellas out of elected office (with the latter being more likely, see Laing and’t Hart, this volume). Leadership transition and succession can also be use-fully considered as social identity processes through which a shift in power(i.e. influence, legitimate authority) takes place.

As we saw earlier, leadership consolidation can be achieved both througha ‘politics of division’ and a ‘politics of solidarity’. Furthermore, these social


identity processes take place across different social identity contexts involv-ing both intragroup and intergroup relations. To understand leadershipchange, therefore, both intra and intergroup factors need to be taken intoaccount. This includes in particular the dynamics of shift towards greater(intra and intergroup) division and conflict, or towards (intra and inter-group) solidarity and cohesion. Figure 10.1 applies this reasoning to under-standing leadership transition and succession. It highlights the distinctionbetween pre- and post-transition/succession contexts. Escalation of inter-group and interparty conflict tends to precede transition, while when it comesto successions such conflict is more likely to be confined to the intragroup,intraparty level.

Let us reflect a bit more on Figure 10.1’s implications. As part of leader-ship succession the ‘politics of division’ are likely to involve leadership con-tests within the party and inter-factional conflict. In contrast, solidarity


Pre-Succession: Within-groupconflict (e.g. intra-partyfactionalism, leadership challenges from within the party)

Post-Succession: Intra-groupunity and cohesion(e.g. party acting as aunited front; ‘up to thejob’)

Post-Transition:Superordinate unity (e.g.‘I am your leader too’;maximizing ‘majority’ inpost-electoral contexts)

Pre-Transition: Inter-groupcontest/conflict (e.g. openinter-party competition inpre-electoral contexts)

Intra-group Inter-group

Division

Solidarity

Transition/Succession

Figure 10.1 Social identity dynamics of leadership consolidation preceding and following transition and succession

emerges through focus on shared party goals and transcending factionalism,as well as a consequence of interparty rivalry. As such, when it comes toleadership succession, the processes of division/solidarity take place at alower, more intragroup, level of analysis. In transition contexts, however, thesame (and at times co-occurring) processes involve higher-order or super-ordinate identities and intergroup relations. The leadership contest is onebetween political parties and therefore focused on highlighting divergentagendas and visions for who ‘we’ are as a nation or state. Post-transition, theemphasis may be on national unity, which can come about as a result of inter-national conflict (e.g. Bush and ‘war on terror), but also through emphasis onthe goals and values that ‘we all share’ (e.g. as in Obama’s presidential victoryspeech).

Transitions thus generally involve highly salient intergroup (e.g. inter-party) boundaries between perennial political opponents. Here, the masspublic is the audience of interest – their malleable but contextually con-strained understandings of the leadership candidates and themselves asgroup members shape the outcome of the interparty contest. In contrast,successions take place on the (more semi-public) intraparty stage. They mayor may not involve an element of interfactional competition within theparty (whether or not there is open competition often depends on the modeof succession, as discussed below). Therefore, shifting a level of analysis islikely to reveal both higher-order intragroup solidarity as part of transitionand lower-level intergroup conflict taking place as part of succession.

To understand more precisely the nature of transition/succession dynam-ics in terms of intra and intergroup division and solidarity, it is necessary toconsider the temporal (pre-, post-) dimension more specifically. As noted,pre-transition and pre-succession contexts are more likely to be marked by(intra and intergroup) division. Transitions are preceded by fierce interpartycontest as part of elections, while successions may involve (equally fierce)inter-factional division over divergent leadership loyalties. In contrast, amove towards greater (intra and intergroup) solidarity is likely to emerge inpost-transition/succession, as parties and government seek to ‘get their acttogether’ and demonstrate to their constituencies that they are indeed‘up to the task’.

Furthermore, pre-transition/succession dynamics involve incumbentleaders seeking to consolidate their own and the party’s position in the faceof potential leadership change – if such power is consolidated successfully,transition/succession may not take place. The weakening of the incum-bent’s influence and authority, while sufficient and perhaps necessary fortransition/succession to take place, may not be enough for the aspiring and‘new’ leaders to gain and maintain influence (Subasic et al., 2009). For post-transition/succession leaders, there is also the important task of craftingand solidifying their psychological bond – a sense of shared identity – withfollowers, both at the ‘elite’ (party, Cabinet) level, but also with their wider


constituencies. The politics of intragroup and intergroup division – and theirrelevant social identity dynamics – are likely to be the hallmark of power con-solidation processes preceding succession and transition, respectively. In con-trast, the politics of solidarity are more prominent immediately followingleadership change.

With the electoral cycle more fully in ‘governing’ mode, (government)leaders may be able to ‘afford’ unpopular and potentially divisive decisions(e.g. Tony Blair’s and John Howard’s 2003 decisions to commit their coun-tries to the Iraq War). Immediately following transition, however, craftinghigher-order shared identities and identity meanings is more central. Forexample, in his victory speech Barack Obama sought to reconcile politicaldivision by explicitly acknowledging Republican Party’s values as thosethat are shared by the nation as a whole:

Let us resist the temptation to fall back on the same partisanship andpettiness and immaturity that has poisoned our politics for so long. Letus remember that it was a man from this state who first carried thebanner of the Republican Party to the White House – a party founded onthe values of self-reliance, individual liberty, and national unity. Thoseare values we all share, and while the Democratic Party has won a greatvictory tonight, we do so with a measure of humility and determinationto heal the divides that have held back our progress. As Lincoln said to anation far more divided than ours, we are not enemies, but friends …though passion may have strained it must not break our bonds of affec-tion. And to those Americans whose support I have yet to earn – I maynot have won your vote, but I hear your voices, I need your help, and I will be your President too (Barack Obama, presidential victory speech,November 2008).

What is also clear from this excerpt, however, is that Obama is using the rhetoric of social unity to distinguish his leadership from that of hispredecessors. In terms of social identity, this is also a process involv-ing exclusion of ‘them’ which solidifies an ‘us’. Here, Obama is expanding the shared identity and maximizing the ingroup ‘majority’ by redefin-ing intergroup boundaries so that the ‘outgroup’ now consists primarily of one’s predecessor and his innermost inner circle (see also Reicher andHopkins, 1996b for a similar approach in the context of Margaret Thatcher’sleadership).

The extent to which leaders can build solidarity and extend one’s sup-port base can also be constrained, however, by the reality of social conflict ininter-(sub)group relations that often takes place both pre- and post-leadershiptransition. In this context, one way to consolidate power is to maximize theextent to which particular messages, decisions and policies can be synchro-nized to appeal (at least on some level) to multiple constituencies. One way to


achieve this is to combine a rhetoric of unity with a clear recognition (ratherthan obfuscation) of subgroup differences. During the 2008 campaign, forexample, Barack Obama’s skilfully handled his opponents’ efforts to createpolitical division along racial lines using such a strategy:

I believe deeply that we cannot solve the challenges of our time unlesswe solve them together – unless we perfect our union by understandingthat we may have different stories, but we hold common hopes; that wemay not look the same and we may not have come from the same place,but we all want to move in the same direction – towards a better futurefor our children and our grandchildren (Barack Obama, ‘A more perfectunion’ speech, March 2008).

Therefore, in these examples the rhetoric of solidarity – the recog-nition of subgroup differences in the context of a broader framework which celebrates and constructively builds on such differences – is being usedboth as a pre-transition strategy to redefine the relevant dimension ofsocial comparison (e.g. class rather than race), but also as a post-transitionattempt at national solidarity that crosses traditional political party divides.

In Australia, the newly elected Prime Minister Kevin Rudd chose the firstact of his leadership to be an official apology to Indigenous Australians.A highly anticipated historical moment for the nation as a whole in termsof Indigenous – Non-Indigenous relations, the apology as noted earlier alsosignalled a clear departure from the previous government’s stance on thishighly divisive issue. The significance of the speech for the arguments pre-sented in this chapter, concerns the way in which Rudd managed to avoidboth ‘assimilationist rhetoric’ (which his predecessor was frequently accusedof) and further division within the Non-Indigenous community concern-ing whether the apology was needed in the first place. Instead, the speechrecognized both the reality of Indigenous – Non-Indigenous relations (e.g.history of oppression and dispossession, disadvantage, etc.), as well as ashared purpose and future aspirations for ‘all Australians’ (including Non-Indigenous Australians):

We today take this first step by acknowledging the past and laying claimto a future that embraces all Australians. A future where this Parliamentresolves that the injustices of the past must never, never happen again.A future where we harness the determination of all Australians, Indigen-ous and non-Indigenous, to close the gap that lies between us in lifeexpectancy, educational achievement and economic opportunity …A future based on mutual respect, mutual resolve and mutual respons-ibility. A future where all Australians, whatever their origins, are trulyequal partners, with equal opportunities and with an equal stake in


shaping the next chapter in the history of this great country, Australia(Kevin Rudd, apology to Indigenous Australians, February 2008).

Like transition, leadership succession is also frequently characterized bythe need to ‘move on’ from the leadership contest within the party andtowards greater party cohesiveness and unity. Seeking to achieve powerconsolidation through intragroup solidarity in the post-succession periodmay have the dual benefit of enhancing a leader’s influence while margin-alizing those within the party who are opposed to change. Post-successionsolidarity is essential to demonstrate to the broader support base that theparty is able to rise above factionalism and act as a united front in face offuture challenges (whether in government or opposition).

Additionally, for ‘new’ post-succession leaders it may also be importantto introduce novel agendas and projects, but in a way that does not furtherfragment the party and enhance factional allegiances. Mode of successionis particularly relevant here as hard-won leadership battles are likely toallow greater scope for innovation and more ‘radical’ policy directions. Itmatters, for example, whether leadership succession is a product of decisivepersonal victories or a matter of coincidence (e.g. Truman and Johnsoninheriting the US presidency from a deceased predecessor). Therefore, whilethe succession ‘stage’ is often an intragroup one (played out within theconfines of the party), it is nevertheless influenced by the leader’s and theparty’s positioning within the broader, intergroup context.

Recent (2009 and 2010) leadership successions within the AustralianLiberal party, for example, were triggered by two factors. Firstly, the viewthat the incumbent leader was no longer able to articulate a shared visionfor the broader constituency and mobilize the public for the party’s agenda(hence the mid-2008 succession from Howard’s successor Brendan Nelsonto the ‘more charismatic’ Malcolm Turnbull). Secondly, the perception thatan incumbent was championing an ‘outgroup’ cause while disregarding‘ingroup’ concerns and interests (hence the late 2009 succession fromMalcolm Turnbull to Tony Abbott, following Turnbull’s failed attempt to get his party behind his willingness to do a deal with the Labor gov-ernment on emissions trading). The second example, in particular, is onewhere the opposition leader’s bipartisan stance – which he attempted to frame in terms of shared, national interests – was seen as a betrayal ofmore narrowly defined ‘ingroup’ or subgroup (the Liberal Party and its coreconstituencies) interests by many of his shadow ministers.

It is also important to recognize, however, that these inter and intra-group dynamics are facilitated and constrained by the relevant institutionalcontext and nature of the office that the leaders inhabit. Political leader-ship takes place both at the ‘elite’ level and the mass-mediated public stage,so that leaders are continually positioning themselves in relation to multi-ple audiences. Contextual and role constraints will shape leaders’ actions in


significant ways. It matters, for example, whether or not a prime minister isleading a multi-party coalition or a single party government. Compared toleading a cohesive, single-party government (e.g. as in the case of JohnHoward and Paul Keating), coalition leadership cannot afford to be as parti-san in terms of content or style. Similarly, the tasks of an incoming oppos-ition leader are different from those of an incoming prime minister. Ongoingneed to campaign and position the party for the next election may be a focusfor an incoming opposition leader, but less of a priority for a new prime minister who may be more focused on the transition from campaigning togoverning.

Similarly, leaders’ relations with the mass public are affected by externalevents and processes that are (at least partly) outside their control. Forexample, at different points in the electoral cycle, leaders have more or lessflexibility to push particular (un)popular policy agendas and messages.Facing the possibility of transition (i.e. elections), both the incumbent andaspiring leadership are in ‘campaign mode’ and will be motivated to resortto messages and identities that mobilize as large a public as possible. In thiscontext, attempts to redefine ‘traditional’ divides (left vs. right, Labor vs.Liberal) abound, as in Liberal Opposition Leader Tony Abbott’s surpriseannouncement to increase taxes for high-profit businesses to fund a gen-erous maternity leave scheme. For Abbott, the prospect of infuriating someof the ‘party faithful’ (i.e. the business community) and his own shadowministers was a risk worth taking given the opportunity to maximize thefemale/traditionally Labor vote across the nation. Conversely, drawing the‘traditional’ divides too sharply is likely to backfire as it is unlikely to securethe ‘majority’ support required in election contexts. Maverick electionticket stalwarts Barry Goldwater or Sarah Palin, for example, galvanized the conservative Republican faithful, but by drawing the ingroup-outgroupbarriers so starkly and narrowly they set themselves and their parties up forelectoral losses.

While significant contemporary events (e.g. a country going into war orfacing a major crisis) present particular leadership challenges at all times,their effect is amplified in leadership change contexts where the need toconsolidate power is paramount. Social creativity strategies are particularlycentral here as power consolidation rests on having the skill to ‘read’ andre-interpret novel events in a way that will resonate with the relevant con-stituency and (simultaneously) discount alternative interpretations. Whichdimensions or aspects of the event are emphasized, what understandingsand emotional responses are suggested as appropriate constitute the pro-cess of (re)defining who ‘we’ are in a given context and how ‘we’ shouldrelate to others. George W. Bush’s response to the events surroundingSeptember 11 and John Howard’s rhetoric surrounding the Tampa crisis(and later the Iraq war, see De Castella et al., 2009) are examples of leaders’ability to successfully leverage contemporary events for particular political


gain by shaping followers views of themselves and others in the relevantsocial context. To consolidate power in leadership change contexts, leadersneed to be able to manage institutional and broader social and historicalconstraints and opportunities in a way that solidifies their position at thecore of who ‘we’ are. Power becomes more difficult to hold on to as rel-evant social identities are redefined in ways that allow alternative sourcesof influence and leadership to emerge and re-cast who ‘we’ are in their ownimage.

Summary and conclusion

Building on current understandings of leadership within social psychologyand the social identity perspective in particular, this chapter focused on theinterplay between intra- and intergroup processes involved in leadershipchange. A novel framework was offered in which power consolidationdynamics involved in leadership transition and succession can be seen as a function of relevant social identity contexts (intergroup, intragroup; inter-party, intraparty) and processes (division, solidarity). The social identity pro-cesses in leadership change contexts (pre- and post-transition, pre- andpost-succession) are not mutually exclusive. Rather they can be considered as(horizontally) inter-dependent, so that division and exclusion simultaneouslyproduce solidarity and inclusion. They are also (vertically) co-occurring, sothere can be parallel divisive and solidary processes occurring at differentlevels of action and analysis, e.g. the party organization vs. the voting public.Finally, we considered how institutional, electoral and socio-historical factorsinteract with social identity processes to shape leadership change.

It could be argued that the politics of solidarity and division are familiarresponses of leaders seeking to consolidate their position at times of change.However, to understand when and why such strategies may help leaders toachieve their desired objectives (and when they may backfire), it is necessaryto move beyond leader-centred approaches and towards an analysis of leader-ship as a group and, importantly, intergroup process. When leadership is seento be about a sense of shared collective meaning and purpose between leadersand followers – where leaders embody ‘our’ shared norms, values and beliefsgiven the social reality of intergroup relations – it becomes possible to under-stand the process by which the politics of division and solidarity serve toconsolidate or undermine one’s influence, authority and power.

Notes

1 Posted on 27 June 2007, the day Gordon Brown succeeded Tony Blair as the PrimeMinister of the UK. Accessed at http://www.youtube.com/watch?v=HEfYI94LzDs.

2 The writing of this chapter was financially supported by funding from the AustralianResearch Council for two Discovery Projects (DP0663314, DP1095319).


Bibliography

Brett, J. (2005) ‘Relaxed and comfortable: The liberal party’s Australia’, Quarterly Essay,19.

Bynander, F. and ’t Hart, P. (2006) ‘When power changes hands: The political psycho-logy of leadership succession in democracies’, Political Psychology, 27(5), 707–30.

De Castella, K., McGarty, C. and Musgrove, L. (2009) ‘Fear appeals in political rhetoricabout terrorism: An analysis of speeches by Australian Prime Minister John Howard’,Political Psychology, 30(1), 1–26.

Dovidio, J. F., Gaertner, S. L., Validzic, A., Matoka, K., Johnson, B. and Frazier, S.(1997) ‘Extending the benefits of recategorization: Evaluations, self-disclosure, andhelping’, Journal of Experimental Social Psychology, 33, 401–20.

Duck, J. and Fielding, K. (2003) ‘Leaders and their treatment of subgroups: Implicationsfor evaluations of the leadership and the superordinate group’, European Journal ofSocial Psychology, 33, 387–401.

Fear, J. (2007) Under the Radar: Dog-Whistle Politics in Australia (Canberra: The AustraliaInstitute).

Gaertner, S. L., Dovidio, J. F., Anastasio, P. A., Bachman, B. A. and Rust, M. C. (1993)‘The common ingroup identity model: Re-categorisation and the reduction of inter-group bias’, European Review of Social Psychology, 4, 3–24.

Giambatista, R. C., Rowe, W. G. and Raiz, S. (2005) ‘Nothing succeeds like succes-sion: A critical review of leader succession literature since 1994’, LeadershipQuarterly, 16, 963–91.

Haslam, S. A. (2004) Psychology in Organizations: The Social Identity Approach, 2nd edn(London: Sage).

Haslam, S. A. and Platow, M. J. (2001) ‘The link between leadership and followership:How affirming social identity translates vision into action’, Personality and SocialPsychology Bulletin, 27(11), 1469–79.

Haslam, S. A., Platow, M. J., Turner, J. C., Reynolds, K. J., McGarty, C., Oakes, P. J.,Johnson, S., Ryan, M. K. and Veenstra, K. (2001) ‘Social identity and the romanceof leadership: The importance of being seen to be “doing it for us”’, Group Processesand Intergroup Relations, 4(3), 191–205.

Kane, J. and Patapan, H. (2008) ‘The neglected problem of democratic leadership’, inP. ’t Hart and J. Uhr (eds) Public Leadership: Perspectives and Practices (Canberra:ANU E Press), pp. 25–36.

Levine, M., Prosser, A., Evans, D. and Reicher, S. (2005) ‘Identity and emergencyintervention: How social group membership and inclusiveness of group bound-aries shape helping behavior’, Personality and Social Psychology Bulletin, 31(4),443–53.

McAllister, I. (2003) ‘Border protection, the 2001 election and the Coalition victory’,Australian Journal of Political Science, 38(3), 445–63.

Pittinsky, T. L. and Simon, S. (2007) ‘Intergroup leadership’, Leadership Quarterly, 18,586–605.

Pittinsky, T. L. (ed.) (2009) Crossing the Divide: Intergroup Leadership in a World ofDifference (Boston, Mass.: Harvard Business School Publishing).

Platow, M. J. and Van Knippenberg, D. (2001) ‘A social identity analysis of leadershipendorsement: The effects of leader ingroup prototypicality and distributive inter-group fairness’, Personality and Social Psychology Bulletin, 27(11), 1508–19.

Rabbie, J. M. and Bekkers, F. (1978) ‘Threatened leadership and intergroup competition’,European Journal of Social Psychology, 8, 9–20.


Reicher, S., Haslam, S. A. and Hopkins, N. (2005) ‘Social identity and the dynamicsof leadership: Leaders and followers as collaborative agents in the transformationof social reality’, Leadership Quarterly, 16(4), 547–68.

Reicher, S., Haslam, S. A. and Rath, R. (2008) ‘Making a virtue of evil: A five-stepsocial identity model of the development of collective hate’, Social and PersonalityPsychology Compass, 2(3), 1313–44.

Reicher, S. and Hopkins, N. (1996a) ‘Seeking influence through characterizing self-categories: An analysis of anti-abortionist rhetoric’, British Journal of Social Psychology,35(2), 297–311.

Reicher, S. and Hopkins, N. (1996b) ‘Self-category constructions in political rhetoric:An analysis of Thatcher’s and Kinnock’s speeches concerning the British miners’strike (1984–5)’, European Journal of Social Psychology, 26(3), 353–71.

Reicher, S. and Hopkins, N. (2001) Self and Nation (London: Sage).Subasic, E., Reynolds, K. J. and Turner, J. C. (2008) ‘The political solidarity model

of social change: Dynamics of self-categorization in intergroup power relations’,Personality and Social Psychology Review, 12(4), 330–52.

Subasic, E., Reynolds, K. J., ’t Hart, P., Reicher, S. D. and Haslam, S. A. (2009a)‘Leadership, social identity and the dynamics of influence in intergroup relations: A new understanding of social continuity and social change’. Unpublished AustralianResearch Council Discovery Project Application, Australian National University.

Subasic, E., Turner, J. C., and Reynolds, K. J. (2009b) ‘How leaders can redefine inter-group relations to shape ingroup identity and consolidate power’. Paper presentedat the International Society for Political Psychology (ISPP) Annual Scientific Meeting,Dublin, 14–17 July.

Subasic, E., Reynolds, K. J., Turner, J. C., Haslam, S. A. and Veenstra, K. (forthcom-ing) ‘Leadership, power and the use of surveillance: Implications of shared socialidentity for leaders’ capacity to influence’, Leadership Quarterly.

Tajfel, H. and Turner, J. C. (1979) ‘An integrative theory of intergroup conflict’, in W. G. Austin and S. Worchel (eds) The Social Psychology of Intergroup Relations(Monterey, California: Brooks Cole).

Turner, J. C. (1982) ‘Towards a cognitive redefinition of the social group’, in H. Tajfel(ed.) Social Identity and Intergroup Relations (Cambridge: Cambridge University Press).

Turner, J. C. (1985) ‘Social categorization and the self-concept: A social cognitivetheory of group behaviour’, in E. J. Lawler (ed.) Advances in Group Processes, vol. 2(Greenwich, Conn.: JAI Press), pp. 77–122.

Turner, J. C. (1991) Social Influence (Buckingham: Open University Press). Turner, J. C. (2005) ‘Explaining the nature of power: A three-process theory’, European

Journal of Social Psychology, 35(1), 1–22.Turner, J. C. and Haslam, S. A. (2001) ‘Social identity, organizations and leadership’,

in M. E. Turner (ed.) Groups at Work: Theory and Research (New Jersey: LawrenceErlbaum Associates), pp. 25–65.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D. and Wetherell, M. S. (1987)Rediscovering the Social Group: A Self-Categorization Theory (Oxford: Blackwell).

Turner, J. C., Reynolds, K. J. and Subasic, E. (2008) ‘Identity confers power: A new viewof leadership in social psychology’, in J. Uhr and P. ’t Hart (eds) Public Leadership:Perspectives and Practices (Canberra: ANU E Press), pp. 57–72.

Van Knippenberg, B. and Van Knippenberg, D. (2005) ‘Leadership self-sacrifice andleadership effectiveness: The moderating role of leader prototypicality’, Journal ofApplied Psychology, 90(1), 25–37.


11Building Trust at the Beginning of aNew Leadership Role: The Role ofLearning and CollaborationPaul Atkins

Entering a new public manager’s role

The nature of public sector management and leadership is changing. Toaddress whole-of-government and whole-of-society challenges, public sectororganizations are exploring more collaborative forms of leadership (Alford andHughes, 2008; Getha-Taylor, 2008). Leadership, particularly at senior levels, isbecoming more about influencing across boundaries, taking broader and moremultiple perspectives and about recognizing and integrating complexity(Hooijberg and Schneider, 2001; Karp and Helgø, 2008). And in public organ-izations in particular, there are increasing demands for transparency and highethical standards from leaders.

This chapter explores how middle to senior managers in the public sectormight begin a new leadership role in such a way that they a) recognize andlearn about the complexity of the challenges they face and b) build trustingcollaborative relationships with staff and stakeholders from the beginning.In complex, multi-stakeholder environments such as those that face mostpublic sector managers, learning and building trust are critical for effective-ness. Subasic and Reynolds (this volume) focus on ways in which creatingclear differentiation between identity groups may aid the new leader indeveloping their power base. In contrast, covering managerial rather thanpolitical successions, this chapter emphasizes processes of integration ratherthan differentiation, focusing on developing trust and learning. It seeks todevelop an approach to succession that is less reactive and more responsivethan that usually adopted by default in most complex organizations, publicas well as private.

The period of succession into a new leadership role is critical for establish-ing trust and credibility in the role of the leader. Just as the new leader cansee with fresh eyes, they also receive heightened attention from their man-agers, peers and subordinates. It is a rare opportunity to learn, both for the

191

leader and for their colleagues and the organization as a whole. Used well,the ‘honeymoon’ period can be a uniquely productive time for examiningold assumptions, creating new alliances and using heightened scrutiny tomodel desired changes in organizational culture (Appelbaum and Valero,2007).

But there are also substantial risks during the time of entry into a newleadership role. This chapter is focused on one of those risks: The tendencyfor new leaders under stress to process information privately rather than collabor-atively. The chapter begins with a review of the situational, cognitive andemotional demands upon a new leader in the middle to senior ranks of thepublic service. It then describes how one natural consequence of thesedemands is a tendency for new leaders to process information privately.After that, the risks and benefits of private and collaborative processing are explored with reference to two key criteria for a successful transition:learning and building trust. A practical, planned approach to commencinga new role is then presented and evaluated in the context of the publicservice.

The demands of a new leadership role

The newly appointed leader in the public service faces significant situ-ational, cognitive and emotional demands. The situation often requires thenew leader solves immediate and pressing operational problems quickly.For example, one of my coaching clients was told during an interview for anew middle management role that she was the ‘silver bullet’ and that she‘had been brought in to fix all that had gone before and to make a dysfunc-tional area functional’. New leaders are naturally proud of their abilitiesand determined to make a positive difference so it is natural to narrowtheir attention to providing immediate solutions.

In many public service agencies, this tendency is further supported by aculture that rewards ‘heroic’ leaders who offer solutions instead of questionsand problem statements. Reward structures and probationary periods aresuch that the most salient task for a new leader is to impress upwards ratherthan build relationships down and out. For example, in the AustralianPublic Service framework, the competency ‘Achieves Results’ is sometimesseen as the only one that really matters (Australian Public Service Com-mission, 2004). Furthermore, in situations with rapid turnover of job roles,there seems little point fostering long-term relationships with staff andstakeholders when relationships are likely to change within a relatively shortperiod.

In short, the context for a new leader in the public service often involvestoo little time, too much to learn, pressing problems created by others butusually with little agreement as to the nature or extent of the problems;uncertainty regarding who to trust; heightened scrutiny from managers,

192 Building Trust at the Beginning of a New Leadership Role

peers and subordinates; limited scope to make decisions independentlywith numerous, complex connecting elements and strong demands to ‘hitthe ground running’.

These contextual factors obviously place significant emotional demandson the new leader. Even if the leader is one who thrives under pressure,significant elements of this context are likely to be perceived as threateningor potentially threatening. In situations of perceived threat, we appear tobe ‘programmed’ to focus attention and act rapidly and automatically(Fredrickson, 2000). This is the so-called fight-flight response, a sophis-ticated combination of cognitive, emotional and somatic reactions leadingto rapid and automatic action to reduce threat (Sapolsky, 1995). From anevolutionary perspective, narrowed attention and rapid reactions are neces-sary to deal with threatening situations. However this pattern of responsemay be sub-optimal in modern complex environments where situationalawareness, informed problem framing and stakeholder relationships arecrucial.

Finally new leaders often face significant cognitive demands. In complexenvironments with multiple elements and non-linear relationships, humanbeings appear to be biased towards rapidly seeking to trial possible sol-utions rather than spending time clarifying and defining problems andgoals (Dörner, 1996). One reason why this occurs is that we wish to protect our sense of competence, and exposing ourselves to the uncertainty andambiguity of complex issues and relationships feels deeply uncomfortable.Furthermore, public servants are not rewarded for professing ignorance andso are likely to rapidly, and even unconsciously, frame issues in such a waythat they have solutions ready to hand. Dörner (1996) refers to this automatictendency to see a problem in terms of known solutions as ‘methodism’.

The following case history illustrates this tendency: John was the newBranch Head of the HR department for a major Australian governmentagency. He came to the position having been acclaimed for a highly success-ful change initiative in a smaller agency where he dramatically increased theamount and scope of training that staff received. After an initial ‘meet andgreet’ with a variety of stakeholders, he noticed clear capability deficits andhe formed the opinion that what was needed was an increase in the amountand variety of training. This opinion was only reinforced following his discussions with the training unit.

John knew he was well placed to use training as a major lever for change.His connections to a variety of training companies meant he knew theindustry and he was himself a fantastic trainer. In the first days of his tenure,he called his staff together and clearly articulated his vision for what couldbe accomplished by improving the skills and capabilities of the entire agency.Although there were some murmurings about all the other work that neededto be done, in general his staff were enthusiastic about the possibilities fornew challenges, getting beyond their procedural work and for engaging

Paul Atkins 193

more with the rest of the agency. In those first days, John quickly estab-lished his authority as an energetic, clear-headed and hard-working leaderoffering a new and exciting direction and identity for the role of HR in theagency.

Unfortunately within six months, the gloss had worn off. Senior manage-ment had received numerous complaints regarding poor handling of basicsystems of leave, pay and recruitment and one particularly prominent staffdispute remained unresolved. John was reassigned to a special project and anew leader was appointed who, in turn, wound back many of the trainingactivities of the HR department and established a ‘back-to-basics’ philo-sophy. Before long the agency was in much the same position as it wasbefore John commenced his role.

So what went wrong? I chose this story because, in some ways at least, Johndid all the ‘right’ things as a new leader: consulting with key stakeholders,identifying a clear and compelling direction for change, and implementinghis vision with energy, expertise and enthusiasm. Below, I argue that John’sresponse to the demands of a new leadership role, while quite typical andnatural given our evolutionary programming, is inadequate for the needs ofthe modern public service, demonstrating how what I call private processingof information can impair both individual and organizational learning.

A common new leader syndrome: Private processing ofinformation

With hindsight it seems obvious John was too strongly influenced by hisprevious experience and did not adapt appropriately to the new situation.New leaders feel great pressure to provide solutions to pressing problems so that they demonstrate their ability and establish their authority. Johnfailed to assist his agency largely because of the way that he paid attentionin the early days of his new leadership: selectively noticing aspects of hisexperience that fitted with solutions he had confidence in offering; dis-counting experiences that did not support his identity as a progressive,knowledgeable leader; failing to engage in a sufficiently collaborative wayto expose and grapple with the inevitable tensions and resource allocationissues afforded by a major change programme; and failing to develop hisown team – leaving them vulnerable to yet another leader with the ‘silverbullet’ for success. In short, John focused and acted on his own private rea-soning and activity without building a collaborative environment to testhis agenda and understanding publicly.

Private processing of information is a form of defensive behaviour thatnegatively impacts upon individual and organizational learning. Emphasizingthe importance of error-detection and correction for organizational learning,Argyris and Schon (1978) distinguished between Model 1 behaviours, whichinvolve conflict avoidance and self-protection; and Model 2 behaviours that


involve relational transparency, public reflection and double-loop learning.People engage in model 1 behaviours, such as concealing their agenda andprivate information processing because they are fearful that being openabout their aims, uncertainties, and information processing may exposethem to conflict, questioning or ridicule. Of course, not all new leaders facethe same demands or react the same way, but the situation and responsesdescribed above are common enough to make it worthwhile to explore theconsequences and what might be done about it.

So far I have argued that the social environment for a new leader in thepublic service often makes substantial cognitive and emotional demandsupon new leaders. These demands include deeply embedded tacit assump-tions that problems and solutions exist somehow independently of the per-spectives brought to them, and that the role of the new leader is to fix thoseproblems as quickly as possible. These assumptions and social demands,combined with natural, evolved patterns of psychological reaction; push theleader towards narrowing their attention and relying heavily upon theirown resources and past experience in order to prove themselves by fixingthings. The result is that new leaders tend to conduct private reasoningregarding:

– The data: in the form of selective noticing and attending; – The nature of the problem: in the form of attributing causes and inferring

the nature of the problem; and– The potential solutions. In the form of creating or replaying previously

successful action plans.

The first two of these, data selection and problem framing are also parti-cularly likely to be conducted ‘on-the-fly’, reactively and automatically. To the extent that these processes occur outside of rational awareness theyare prey to systematic distortions such as the availability bias (Tversky andKahneman, 1974), confirmation bias (Gilbert, 1991) and defensive routinesto avoid uncertainty, conflict and perceptions of incompetence (Dörner,1996).

The demands outlined above can be seen as a set of tacit short-term criteriafor a successful succession. But if we broaden our focus to consider how anew leader might make progress on complex organizational challenges inthe longer term, a different set of criteria come into focus. These long-termcriteria are relevant even in the first days of succession because patterns ofinteraction tend to persist and the first days of a new role are the bestopportunity the leader will get for learning. Figure 11.1 represents possiblecriteria for a successful succession, from the perspective of the leader, in termsof an inherent tension between short- and long-term interests.

The urgency of the short-term demands means that they dominate theagenda of a new leader unless (s)he engages in a conscious and deliberate

Paul Atkins 195

strategy for entering the role. I explore such a deliberate strategy towardsthe end of this chapter. But in the following section I aim to make the casethat leaders can and should consider the long-term criteria even in the faceof the short-term demands of succession.

An alternative: Building trust through authentic leadership

Trust is essential for long-term team performance and successful stakeholderengagement. When followers trust their leaders, they perform better on job-relevant tasks, engage in more organizational citizenship behaviours, experi-ence greater job satisfaction, display greater organizational commitment,and are less likely to wish to leave the organization (Dirks and Ferrin, 2002).Furthermore, team and organizational performance is improved (Burke etal., 2007) and followers are more likely to be satisfied with their leaders andperceive their leader to be effective (Gillespie and Mann, 2004).

So how is trust in a leader developed? There is now good empirical evid-ence on this question, and the answer is complex. In a multi-level review,Burke et al. (2007) pointed to a range of antecedents and moderators oftrust in leadership. Antecedents included a consultative leadership style,behavioural integrity (consistency of word and deed), leader ability, andbenevolent behaviour; while moderators included leader reputation as wellas trustor, team and organizational characteristics. Trust is developed throughbeing capable, conveying legitimacy and purpose, doing what one says onewill do and involving others in decisions.

Of these, most new leaders seem to tacitly focus upon demonstratingtheir ability and bolstering their reputation in order to build trust. In anenvironment where they do not know who to trust and how much, ratherthan depend upon others new leaders seek modes of action that rely firstand foremost upon their own capabilities. One way they can do this is to


stakeholders.

Short-term

1. Dealing effectively withimmediate operationaldemands and issues (‘hit theground running’)

2. Establish credibility bycreating perceptions ofcompetence and legitimacy(particularly up, but also downand out)

Long-term

1. Building trustingrelationships (particularlydown but also up and out)

2. Managing self to worksustainably (maintainingengagement and minimizingburnout)

3. Enabling strategic individualand organizational learning

Figure 11.1 Short-term demands versus long-term interests. ‘Up’ refers to one’smanagers, ‘down’ refers to one’s team while ‘out’ refers to peers and stakeholders

dive into solving pressing and complex problems. Their capability is thusdemonstrated through immediate achievement. But this is a high-risk stra-tegy, which often sees the leader working on the wrong issues or ignoringvaluable contributions from others in the process. In that case, efforts atunilateral problem solving result in trust evaporating rather than beingbuilt. In a connected world, we need more sustainable and reliable ways toestablish leadership authority and trust.

One factor that appears to be critical in doing so is a consultative leadershipstyle. Involving followers in participative or at least consultative decision-making generates perceptions of organizational justice and trust in the leader(Dirks and Ferrin, 2002). Indeed, Gillespie and Mann (2004) found that con-sultation with team members was one of three key factors (along with com-municating a vision and establishing shared values) that accounted for 67 percent of the variance in team members’ trust towards their leaders. Consult-ative leadership enhances perceptions of autonomy which, in turn, enhancesmotivation and engagement in followers (Stone et al., 2009). Furthermore,consultation helps develop a clear sense of shared purpose and empowers fol-lowers to realize that purpose (Burke et al., 2007). More broadly, the notion ofrelational transparency includes but extends the consultative leadership style.According to Mazutis and Slawinski (2008: 445) relational transparency:

[E]ncompasses all of the earlier capabilities in the act of open and truth-ful self-disclosure … In addition to being self-aware, balanced and con-gruent in one’s goals, motives, values, identities and emotions, authenticleaders are also transparent in revealing these expressions to their followers(Hughes, 2005). Disclosing one’s true self to one’s followers builds trust and intimacy, fostering teamwork and cooperation (Gardner et al., 2005)and feelings of stability and predictability (Chan et al., 2005). Furthermore,relational transparency requires the willingness to hold oneself open forinspection and feedback, thereby also being an essential component in thelearning process (Popper and Lipshitz, 2000).

In other words, relational transparency involves risking vulnerability byrevealing one’s own mental and emotional processing for public scrutiny.

The consistency between the leader’s word and deed, referred to as behavioural integrity, also appears to be critical for trusting a leader andmay, in turn, lead to enhanced performance, satisfaction and organ-izational citizenship (Simons, 2002). Palanski and Yammarino (2009) used evidence from studies of transformational (Bass and Steidlmeier,1999), authentic (Avolio and Gardner, 2005) and spiritual (Fry, 2003) leadership, which all feature behavioural integrity, to argue that behav-ioural integrity should positively affect at least the social aspects of leader performance (e.g. consulting stakeholders, managing conflict, giving feedback).

Paul Atkins 197

Behavioural integrity builds trust because followers can more confidentlypredict the leader’s behaviour. But such confidence arises from a pattern ofconsistent behaviour over time. To be consistent in the face of a changingand demanding environment, the leader needs to be able to effectively self-regulate (Atkins, 2008). Self-regulatory capability is what allows us to predictour behaviour and to regulate our internal and external state in order toachieve valued ends. So to act with integrity in a changing complex envi-ronment, leaders require self-awareness, as manifested by an ongoing,dynamic appreciation of their own sense-making in context. In other words,leaders need to be able to take a perspective on their own values, emotions,identity and characteristic behavioural reactions in relationship to theirexperience of the world (Kegan, 1994).

One recent approach to leadership development that has brought toge-ther these elements of relational transparency, behavioural integrity andself-awareness is that of authentic leadership (Avolio and Gardner, 2005).Gardner et al. (2005) define authenticity as ‘both owning one’s personalexperiences (values, thoughts, emotions and beliefs) and acting in accord-ance with one’s true self (expressing what you really think and believe andbehaving accordingly)’. More specifically, authentic leadership has beencharacterized in terms of four factors: self-awareness, authentic behaviour,balanced processing and relational transparency. Self-awareness includesnot just awareness of the content of one’s experience but also of the epi-stemological process of framing and constructing that experience. Authenticbehaviour is similar to behavioural integrity in that it involves a consist-ency of word and deed. But it also includes the idea of being true to one-self. According to Kernis (2003: 14) ‘behaving authentically means acting inaccord with one’s values, preferences, and needs as opposed to acting merelyto please others or to attain rewards or avoid punishments’. In this sense,authentic behaviour reflects intrinsic motivation and self-determination (Deciand Ryan, 1985, 2008).

Balanced processing refers to being able to rationally process one’s ownand others’ perspectives. It is about being open to self-relevant informationwithout ego-defence. Authentic people ‘are able to more objectively evalu-ate and accept both positive and negative aspects, attributes and qualitiesof themselves [and others], including skill deficiencies, suboptimal per-formance, and negative emotions’ (Gardner et al., 2005: 356). This non-defensive and open awareness takes time to develop but it is central toeffective dialogue and collaboration. Indeed, there is increasing evidencethat this form of awareness, often referred to as mindfulness, is predictiveof a wide range of positive health and performance outcomes (Brown et al.,2007).

Recently a measure of authentic leadership has allowed researchers to start exploring the impacts of this specific constellation of leader activities.Walumbwa et al. (2008: 111) found that their measure ‘accounted for


variance in a diverse set of frequently researched work outcomes beyondthat explained by ethical and transformational leadership dimensions. Spe-cifically, authentic leadership was shown to account for additional uniquevariance in OCB [organizational citizenship behaviours], organizationalcommitment, and satisfaction with supervisor. Thus, evidence was obtainedof the value added by considering the effects of authentic leadership on com-monly researched organizational variables beyond existing measures of relatedleadership constructs’. In addition, ‘follower perceptions of the leaders’authentic leadership were positively related to individual follower job satis-faction and rated job performance, controlling for the effect of organizationclimate’ (Walumbwa et al., 2008: 117). Authentic leadership appears to pos-itively impact performance, job satisfaction as well as leader and followerwellbeing (see Avolio et al., 2009 for a review). Furthermore, trust in leader-ship appears to at least partially mediate the relationship between authenticleadership and performance (Clapp-Smith et al., 2009).

In summary, in this section I have reviewed evidence regarding the impor-tance of trust in achieving positive work outcomes and discussed a form ofleadership that appears to contribute to trust and performance. In terms ofthe demands of succession, I am suggesting that a leader might not onlyattend to the immediate demands, but could also consider what types of relationships they wish to develop in order to be most effective in thelonger term. A leader who begins with private processing of informationand unilateral action may rapidly resolve initial crises but the long-runimpact may include undermining trust or, at the very least, disempoweringfollowers to contribute at their best. Expectations, both positive and neg-ative, become entrained such that early impressions influence later inter-pretations of leader actions.

Clearly trust and open communication facilitate organizational and individual learning. It is extremely easy for a person in authority to becomeisolated from the true opinions and perspectives of their followers andpeers. But behaving in this collaborative way is by no means easy. Leadersoften need to learn new ways of seeing their role and value, as well as moreeffective ways to express and regulate emotions. In the next section, I des-cribe in detail one potential way in which a structured approach might beused to deliberately support a more collaborative and open form of engage-ment with staff.

Enabling strategic individual and organizational learning: Astructured approach to collaborative enquiry for the new leader

In response to the challenges of a new leadership role Jentz (2008) pro-posed a process called ‘EntryPlan’ to strategically orient new leaders to theirenvironment and establish credibility and trust up, down and outwards.The EntryPlan approach was designed to encourage new leaders to select

Paul Atkins 199

data and define problems more consciously and publicly. In this section I briefly describe this approach and evaluate it in the light of the researchdescribed earlier. An EntryPlan comprises four stages: a) designing andpublic testing, b) gathering data, c) conducting joint sense-making, and d)collaborative action planning. Each of these stages is described in briefbelow and in detail by Jentz (2008). Importantly, this work does not occurinstead of working on immediate and pressing problems, rather it is expectedthat these steps will be conducted as one project among others that need tobe accomplished in the early days of a new leader’s role.

Many new public leaders are, at least initially, primarily concerned withimpressing those who appointed them. In my experience, it is importantthat the EntryPlan approach is endorsed by one’s managers. The processappears to be most effective if the new leader sets expectations early thatthey are going to engage in a structured succession process that is designedto foster trust and learning. This process could start with the interview for the position, where having an EntryPlan is likely to be seen as a strongindicator of strategic thinking and commitment to the position.

Stage 1: Design and test the EntryPlan

Key decisions at this stage are: a) what is the purpose of the EntryPlan,b) who to talk to, and c) what questions to ask. The design can be con-ducted individually, with a manager or coach depending on the leadercharacteristics and the extent and nature of the demands of the leadershiprole.

Designing the plan: What is the purpose of the EntryPlan?

Two common reasons for engaging in this process are to orient to thestrategic environment and to build relationships within one’s own team.All leaders both set an agenda and respond to others’ agendas. Leadershipis a relationship between productive and receptive forces. But at seniorlevels, setting the agenda and deciding on strategic directions becomes adominant responsibility over and above line management and operationaltasks (Australian Public Service Commission, 2004). In the words of oneExecutive Director reflecting on his transition from being a Branch Head:‘the focus on [task expertise and execution] dissipates and the expertisebecomes more about stakeholder relationships and inter-governmentalstuff and building that overall strategic knowledge into the organization’.

Designing the plan: Who needs to be included in what steps of theprocess?

At the more junior levels, leaders may focus on establishing relations withtheir direct reports during transition into the role. As the role becomesmore senior, it is possible that the emphasis will shift from direct reports tostrategic external stakeholders. This depends on what the leader hopes to


achieve. Who do they most need to establish trust and credibility with?Who has the most critical perspectives for them to complete their workeffectively? Stakeholders can be selected depending on their interest orpower in the tasks relevant to the leader’s role. While a first draft might becompleted by the leader themselves, it is likely that the plan will go througha number of iterations including consultation with more senior leaders todetermine key stakeholder relationships.

Designing the plan: What questions to ask?

The scope and focus of the plan will be determined by its purpose. Questionsmight focus on areas such as: a) issues and principles that matter most to theinterviewee, b) what the interviewee might know that could help the leadersucceed (or avoid failure), c) current issues facing the interviewee’s organ-ization/unit and what is on the horizon and d) how the different areas mighteffectively work together.

Testing the plan

Finally, in the spirit of collaborative reasoning, Jentz (2008) argues that theplan itself should be made public to others such as the leader’s manager,their team or other key stakeholders who might be able to make useful con-tributions to its structure.

Stage 2: Gathering the data through dialogue

Within the EntryPlan process, initial data is usually gathered in structuredand purposeful individual interviews with key stakeholders. This is quitedifferent way to the usual ‘meet-and-greet’ sessions common within thepublic service which often do not have an explicit agenda beyond ‘gettingto know you’ and are fairly unhelpful. Even where they do have an explicitagenda, they are rarely cast in such a way as to focus upon the differentperspectives brought to the issues. Within the EntryPlan approach, datagathering has the qualities of dialogue rather than discussion:

a) Wherever possible, the agenda is open. Purposes are explicitly stated and checked with the other. (e.g. ‘I would like to use this meeting to … better understand your perspective on key challenges and strategicimperatives; … to let you know how I am thinking about my role; … toexplore more effective ways for our groups to work together … How isthat for you?’).

b) Conversation is conducted in such a way that, as much as possible:i. The data (what actually happened or is happening) are kept separate

from interpretations of those data (e.g. ‘In the past our departmentshave often not shared information’ is a much more useable fram-ing of an issue than ‘in the past our departments have concealedinformation from one another.’ The latter infers negative intent that

Paul Atkins 201

reduces possibility of other interpretations and usually gets in theway of building relationships.)

ii. Interpretations are recognized as subjective rather than objectivetruths.

iii. The interviewer’s interpretation of the interviewee’s comments ismade open for testing and further elaboration (e.g. ‘So it sounds tome as though the biggest issue you are noticing here is X, and youwant Y, have I got that right?’)

So in the EntryPlan approach, interviews have an explicitly stated purposeand are conducted using reflective listening skills closer to dialogue thansimple discussion. Conversations that proceed along the lines of: ‘What do you think?’ and then ‘That’s interesting, here is what I think’ do nottypically reveal private assumptions and automatic processing.

Stage 3: Collaborative sense-making

The last two stages of the EntryPlan approach are making sense of the data and prioritizing actions. In some situations, the new leader may plan to simply conduct the data gathering publicly and then process that information privately or with selected colleagues. However, this is a lostopportunity to use the transition into a new role to demonstrate relationaltransparency and bring stakeholders together in a new and productive way.A middle manager who has focused their plan on conversations with directreports could, for example, generate significant engagement, learning andtrust by conducting sense-making and planning meetings with their team.

Collaborative sense-making involves the leader presenting back the sensethey have made of the data gathered in stage 2 for further comment, inte-gration and elaboration. This allows the leader to validate their understand-ing and acknowledge inputs from others. In my experience, collaborativesense-making works best if the leader is as transparent as possible and able to separate ‘facts’ from interpretations. It is at this point that thedefensive behaviours mentioned above are most likely to occur. For leaders unaccustomed to genuine dialogue, this stage can feel extremely uncom-fortable as what previously looked simple in their own minds becomes acomplex web of potentially competing perspectives. To engage fully withthat complexity seems to require the self-regulatory skills discussed earlier.However, exposing this complexity can both inform the leader and make itclear to other stakeholders what constraints the leader is under.

Stage 4: Collaborative Action Planning

While the EntryPlan approach is specifically designed to encourage activeand more deliberate ‘sense-making’, in order to make an impact thereneeds to be a translation into an action plan. In many public sector agen-cies, it seems that developing a prioritized action plan is the sole preserve


of the leader (after all, what are they paid for?). But again, even if action-planning is not done in a group, it can be conducted more collaborativelyby, for example, developing a draft plan which is circulated for commentand input and may be integrated with a wider strategic plan.

So, in summary, while it is possible to conduct only the first two stages,conducting all four publicly and collaboratively could potentially generatevalue by a) improving the quality of information used by the leaders tomake decisions, b) improving the quality of problem definition, c) increas-ing trust in the leader, d) setting expectations that others can and shouldcontribute to solutions, e) increasing the ability of stakeholders to perceivecomplexity and develop experience and skills in working with others tofigure out what to do in the interest of improving the organization,f) increasing shared understanding of current practice and what needs tochange, and g) creating increased engagement through shared perceptionsof the issues and potential action plans (Jentz, personal communication).

The process proposed by Jentz (2008) is potentially a ‘recipe’ for increasingrelational transparency and, consequently, trust and learning as describedearlier. Of course, there is a considerable amount of skill involved in con-ducting this process effectively and these skills appear to be the same as thoseassociated with authentic leadership as described previously. The very act ofengaging in learning about and conducting this process is, therefore, a poten-tially potent form of leadership development in its own right.

Possible objections and risks

In this chapter I have suggested that key tasks of a new leader includedeveloping trust and learning as well as getting started on immediate work.Studies of leader trust and authentic leadership suggest that a collaborativeand transparent approach is likely to be helpful. The EntryPlan process pro-posed by Jentz (2008) provides a structure within which a suitably skilledor supported leader could possibly avoid the pitfalls of private informationprocessing and embody collaboration in the early days of their leadership,thereby building trust and learning.

In my experience, the collaborative approach presented in this chapter is sometimes seen as a bit unrealistic for the busy lives of public servants.And indeed, the importance of sensitivity to the specific demands of theleader’s role cannot be underestimated. Sometimes, for example, a new leadersimply will not have the time or authorization to engage in anything likethe process discussed above. On the other hand, any new approach is likelyto feel uncomfortable, and in this case this problem is particularly acutebecause it asks leaders to make use of self-regulatory skills that can be dif-ficult and uncomfortable to learn. However, given appropriate placementduring interviews and managerial support, the EntryPlan can be framed asan initial project that is part of the work. This view only makes sense when

Paul Atkins 203

the ‘real work’ of leadership is seen as more than just providing solutions inthe short term. If framing the right issues and developing the trusting rela-tionships necessary for effecting change are also valued, then activities likecollaborative sense-making make sense. As a Branch Director noted to me:‘when you really need something from one of the other agencies, if youhaven’t got those relationships built up, you just can’t get it’.

In some ways, the environment of the public service is becoming increas-ingly supportive of collaborative approaches to succession. The push for ‘whole-of-government’ collaboration within public service agencies provides an excellent context for considering collaborative succession as part of the real work of a public leader. It is expected that senior leaders build relationships across agencies and seek ways of seeing strategicissues from multiple perspectives. But often, the whole-of-governmentmessage is largely aspirational, and lacks an actionable plan for indi-vidual leaders. A collaborative approach to succession like that outlinedabove offers an unparalleled opportunity to integrate across organizationalboundaries.

Some leaders express concern about their capacity to be open about theirreasoning and the collaborative approach does indeed call for more trans-parency than most of us are used to. Transparency can also be damaging ifapplied inappropriately for the context. It is obviously easiest to be open inhigh-trust situations and sometimes it is necessary to conceal aspects of ourreasoning in low-trust situations. However, most of us systematically con-ceal aspects of our reasoning even in neutral situations because of largelypersonal reasons such as fear of appearing incompetent or naïve, givingaway power or offending people. This chapter has suggested that leadersare likely to have more success building trust and learning if they are ableto collaborate openly even in the face of these negative emotions. The evidence sug-gests that the most admired leaders are those who are willingto expose their reasoning to scrutiny and act with integrity (Kouzes andPosner, 2006).

Unfortunately, the cultures of many public service agencies do not sup-port open sharing of information. One public manager described her experi-ence of trying to engage in more open information sharing within the taxoffice as follows: ‘sharing knowledge was an anathema to the way they oper-ated, knowledge is power, so [sharing knowledge] was a threat’. Anotherexpressed scepticism that she could be open about her intent to introduce anew compliance process for a key stakeholder. She felt that discussing thisprocess with the stakeholder might trigger such strong reactions in the otherthat the relationship would be destroyed. But the relationship wouldarguably have been destroyed anyway when the stakeholder learned theprocess had been introduced without their knowledge. Furthermore, thereare ways of framing a conversation to allow even the most difficult topics tobe discussed.


For example, continuing this example, the following passage illustrateshow a leader might skilfully elicit the stakeholder perspective in this difficultsituation:

‘The intention of this system is to improve services to clients and publicaccountability. It seems to me highly likely that this system or some-thing like it will be introduced. It is in my interest to make sure that allthe key players are reasonably satisfied with the introduction of thissystem because your cooperation will be critical to its long-term success.I can see merits in the system and one way we could approach thiswould be for us to have a debate about it with me presenting my argu-ments and you presenting yours. But this is unlikely to end in a solutionthat is mutually acceptable because it is unlikely each of us will be ableto set aside our own views enough to fully understand the other’s per-spective. So I would like to try something different and use this meetingto suspend arguing my perspective for now so that I can fully listen to yours. What I would like for this meeting is, as best I can, to learn tosee the issue “through your eyes”. So I intend to listen as closely as I can to your perspectives and to check my understanding as I learn more.I would greatly appreciate it if you were to tell me when you think I amnot appreciating your perspective as fully as I might. How does thatsound to you?’

Assuming the leader is then able to reflectively listen to the stakeholder,they are then in a stronger position to request the same respectful attentionto their perspective.

In conclusion

Faced with the need to demonstrate their ability and establish their author-ity, tacit models of ‘heroic’ leaders who have all the answers seduce newleaders into thinking that private processing of information and unilateralaction is the way to accomplish these aims. This tendency is further exacer-bated by a natural tendency to narrow attention in reaction to stressful situations. Unfortunately, the complex issues faced in most public serviceagencies are such that private reasoning regarding critical problems is aluxury that can no longer be afforded. If we are to make progress on ourmost pressing and systemic issues, we are going to have to learn to face the discomfort associated with exposing our reasoning to scrutiny andthereby communicating and collaborating more effectively. The process of succession into a new leadership role is a unique opportunity to developand demonstrate the possibilities arising from more public and careful reasoning.

Paul Atkins 205

Bibliography

Alford, J. and Hughes, O. (2008) ‘Public value pragmatism as the next phase of publicmanagement’, American Review of Public Administration, 38(2), 130–48.

Appelbaum, S. H. and Valero, M. (2007) ‘The crucial first three months: An analysisof leadership transition traps and successes’, Journal of American Academy of Business,11(1), 1–8.

Argyris, C. E. and Schon, D. (1978) Organizational Learning: A Theory of Action Perspective(Reading, Mass.: Addison-Wesley).

Atkins, P. W. B. (2008) ‘Leadership as response not reaction: Wisdom and mindful-ness in public sector leadership’, in P. .t Hart and J. Uhr (eds) Public Leadership:Perspectives and Practices (Canberra: ANU E Press), pp. 73–82.

Australian Public Service Commission (2004) ‘The integrated leadership system’.Accessed at http://www.apsc.gov.au/ils/index.html.

Avolio, B. J. and Gardner, W. L. (2005) ‘Authentic leadership development: Gettingto the root of positive forms of leadership’, Leadership Quarterly, 16(3), 315–38.

Avolio, B. J., Walumbwa, F. O. and Weber, T. J. (2009) ‘Leadership: Current theories,research, and future directions’, Annual Review of Psychology, 60(1), 421–49.

Bass, B. M. and Steidlmeier, P. (1999) ‘Ethics, character, and authentic transforma-tional leadership behavior’, Leadership Quarterly, 10(2), 181–217.

Brown, K. W., Ryan, R. M. and Creswell, J. D. (2007) ‘Mindfulness: Theoretical foundations and evidence for its salutary effects’, Psychological Inquiry, 18(4),211–37.

Burke, C., Sims, D. E., Lazzara, E. H. and Salas, E. (2007) ‘Trust in leadership: A multi-level review and integration’, The Leadership Quarterly, 18(6), 606–32.

Chan, A., Hannah, S. and Gardner, W. (2005) ‘Veritable authentic leadership:Emergence, functioning, and impacts’, in W. Gardner, B. Avolio and F. Walumbwa(eds) Authentic Leadership Theory and Practice: Origins, Effects and Development,Monographs in Leadership and Management, Volume 3 (Boston: Elsevier).

Clapp-Smith, R., Vogelgesang, G. and Avey, J. (2009) ‘Authentic leadership and positivepsychological capital: The mediating role of trust at the group level of analysis’,Journal of Leadership and Organizational Studies, 15(3), 227–40.

Deci, E. L. and Ryan, R. M. (1985) ‘The general causality orientations scale: Self-determination in personality’, Journal of Research in Personality, 19(2), 109–34.

Deci, E. L. and Ryan, R. M. (2008) ‘Self-determination theory: A macrotheory ofhuman motivation, development, and health’, Canadian Psychology/Psychologiecanadienne, 49(3), 182–5.

Dirks, K. T. and Ferrin, D. L. (2002) ‘Trust in leadership: Meta-analytic findings andimplications for research and practice’, Journal of Applied Psychology, 87(4), 611–28.

Dörner, D. (1996) The Logic of Failure: Recognizing and Avoiding Error in Complex Situations(Cambridge, Mass.: Perseus).

Fredrickson, B. L. (2000) ‘Why positive emotions matter in organizations: Lessonsfrom the broaden-and-build model’, Psychologist-Manager Journal, 4(2), 131–42.

Fry, L. W. (2003) ‘Toward a theory of spiritual leadership’, Leadership Quarterly, 14,693–727.

Gardner, W. L., Avolio, B. J., Luthans, F., May, D. R. and Walumbwa, F. (2005) ‘“Canyou see the real me?” A self-based model of authentic leader and follower develop-ment’, Leadership Quarterly, 16(3), 343–72.

Getha-Taylor, H. (2008) ‘Learning indicators and collaborative capacity: Applyingaction learning principles to the U.S. Department of Homeland Security’, PublicAdministration Quarterly, 32(2), 125–46.


Gilbert, D. T. (1991) ‘How mental systems believe’, American Psychologist, 46(2),107–19.

Gillespie, N. A. and Mann, L. (2004) ‘Transformational leadership and shared values:The building blocks of trust’, Journal of Managerial Psychology, 19(6), 588–607.

Hooijberg, R. and Schneider, M. (2001) ‘Behavioral complexity and social intelligence:How executive leaders use stakeholders to form a systems perspective’, in S. J. Zaccaroand R. J. Klimoski (eds) The Nature of Organizational Leadership: Understanding thePerformance Imperatives Confronting Today’s Leaders (San Francisco: The Jossey-Bassbusiness and management series), pp. 104–31.

Hughes, L. (2005) ‘Developing transparent relationships through humor in the authenticleader–follower relationship’, in W. Gardner, B. Avolio and F. Walumbwa (eds) AuthenticLeadership Theory and Practice: Origins, Effects and Development, Monographs in Leadershipand Management, Volume 3 (Boston: Elsevier).

Jentz, B. (2008) How to Begin a Leadership Position Successfully: The Entryplan Approach(Newton, Mass.: Leadership and Learning Inc).

Karp, T. and Helgø, T. I. T. (2008) ‘From change management to change leadership:Embracing chaotic change in public service organizations’, Journal of Change Manage-ment, 8(1), 85–96.

Kegan, R. (1994) In Over Our Heads: The Mental Demands of Modern Life (Cambridge,Mass.: Harvard University Press).

Kernis, M. H. (2003) ‘Toward a conceptualisation of optimal self-esteem’, Psycho-logical Inquiry, 14, 1–26.

Kouzes, J. and Posner, B. (2006) ‘It’s not just the leader’s vision’, in F. Hesselbein and M. Goldsmith (eds) The Leader of the Future 2 (San Francisco: Jossey-Bass), pp. 207–14.

Mazutis, D. and Slawinski, N. (2008) ‘Leading organizational learning through authenticdialogue’, Management Learning, 39(4), 437–56.

Palanski, M. and Yammarino, F. (2009) ‘Integrity and leadership: A multi-level conceptual framework’, Leadership Quarterly, 20(3), 405–20.

Popper, M. and Lipshitz, R. (2000) ‘Organizational learning: Mechanisms, cultureand feasibility’, Management Learning, 31(2), 181–96.

Sapolsky, R. M. (1995) Why Zebras Dont Get Ulcers: A Guide to Stress, Stress-RelatedDiseases, and Coping (New York: W.H. Freeman and Company).

Simons, T. L. (2002) ‘Behavioral integrity: The perceived alignment between managers’ words and deeds as a research focus’, Organization Science, 13, 18–35.

Stone, D. N., Deci, E. L. and Ryan, R. M. (2009) ‘Beyond talk: Creating autonomousmotivation through self-determination theory’, Journal of General Management,34(3), 75–91.

Tversky, A. and Kahneman, D. (1974) ‘Judgment under uncertainty: Heuristics andbiases’, Science, 185(4157), 1124–31.

Walumbwa, F., Avolio, B., Gardner, W., Wernsing, T. and Peterson, S. (2008) ‘Authenticleadership: Development and validation of a theory-based measure’, Journal ofManagement, 34(1), 89–126.

Paul Atkins 207

12‘The Tideless Pond that SeemedWaiting for Me’:1 The Afterlife ofAustralian Prime MinistersPaul Strangio

After power has changed hands

When power changes hands there are winners and losers: those who gainand those who relinquish power. Much of this volume is oriented towardsthe former: the acquiring of, and adaptation to, government and leadershipoffice. This chapter, however, focuses on what happens on the down sideof the exchange of power. It explores the ‘afterlives’ of chief executives – thetrajectories of those who have surrendered the reins of national governmentin Australia. With the exception of the United States (Norton Smith andWalch, 1990; Skidmore, 2004; Updegrove, 2006), systematic studies of thefate of former leaders in representative democracies are rare.

In a recent meditation on the topic, the political historian, John Keane,observed that the ‘subject of ex-office holders is under-theorized, under-researched, under-appreciated’ (2009: 282–3). For Keane, this neglect is mir-rored in a larger ‘odd silence about life after high office’ (2009: 281) – ‘odd’since a fundamental innovation of representative democracy was its insist-ence that power is transitory and the holders of executive office politicalmortals. Former leaders are, in other words, inherent to properly function-ing democracies in which there is regular rotation of power. But Keane alsopoints to a developing trend of ex-office holders becoming more ubiquitousand active as a result of such factors as improved health and extended lifespans. As ‘[public] life after office is becoming commonplace’ (2009: 282),he suggests former leaders are a source of both problems and opportunityfor democracies.

This chapter takes up Keane’s injunction for serious treatment of ex-leaders:for a better understanding of the changing dynamics that govern the choicesavailable to former heads of government and the nature of the public after-lives they create. Whereas Keane’s is a free-wheeling, conceptually-driven trans-national discussion, here the focus is on the evolution of prime ministerial

208

afterlives in Australia on the premise that situation-specific, empirically-based inquiry is a prerequisite to filling the lacunae of post-leadership research.At the same time, this case study is not sui generis – changes to post-leadershippaths in Australia parallel trends highlighted by Keane, as well as by KevinTheakston (2006) in a recent overview of the activities of former occupantsof 10 Downing Street. Australia is certainly analogous in that the phenom-enon of life after national leadership has escaped scrutiny, apart from someimpressionistic media-based commentary on the current crop of living ex-PMs. The thrust of this commentary (e.g. Gold, 2007; Ray, 2008; Switzer,2008; Warhurst, 2008) paints former office-holders as a meddlesome, largelypurposeless group who have struggled to accommodate themselves to lossof power and of whom it would be better, particularly for their successors,if they were never heard from again. A former senior minister has summedup the plight of the nation’s ex-leaders this way:

prime ministers rarely find peace having relinquished office. They wanderaround aimlessly, seeking out a meaningful role … Mostly, they suc-cumb to generational jealousy, criticising their successors with maximum intensity, especially if they are from the same side of politics (Ray, 2008).

Does the history of prime ministerial afterlives in Australia bear out such aninterpretation? Has it been the inevitable fate of former PMs to becomedisenchanted with heirs and to suffer from office-dependency? Is it true, asKeane suggests (2009: 284), that giving up power ‘gracefully’ does not comenaturally: ‘stepping down is a capacity learned reluctantly, and with thegreatest difficulty’? More fundamental to the question of how powerchanges hands: have former leaders’ wielded power beyond the politicalgrave? In other words, despite their formal relinquishment of office, whatopportunities have they had to continue to exercise influence over succes-sors, their political parties and the public, and in what ways have thoseopportunities altered over time?

Australia’s former prime ministers: An overview

This study delineates a significant change in post-prime ministerial trajec-tories in Australia that can be separated into two distinct eras. It posits the1966 exit of Robert Menzies (the nation’s long-serving PM) as a transitionpoint between those eras. First, though, let us lay out some preliminarycontext and data to undergird this investigation.2 The office and powers ofthe prime minister in Australia are rooted in Westminster-derived customsof responsible government. Traditionally the leader of the majority party inthe House of Representatives, PMs do not formally possess power in theirown right – cabinet is the supreme site of (collective) decision-making. Yetthe classic formulation of PMs as primus inter pares is a long-standing myth.

Paul Strangio 209

As the doyen of scholars of Australian executive government has noted, PMs ‘arethe dominant figures in Australian politics and have always been’ (Weller, 2007:460). Pivotal to this dominance has been a stable party system (Labor versus non-Labor) that congealed at the end of the first decade of the Commonwealth(Strangio, 2009). Whereas before that settlement there had been a rapid turnoverof ministries most of which were terminated on the floor of the parliament, inthe hundred years since, most PMs have lost office because of electoral defeat. Indescending order of frequency, other causes have been overthrow by parliamen-tary subordinates, death (Joseph Lyons, John Curtin and Harold Holt), retire-ment, defeat in the House of Representatives and dismissal. It has become anaxiom of Australian politics that, once elected, governments can expect at leasttwo terms of office – the last administration not returned on its first visit to thepolls was James Scullin’s Depression-era Labor ministry, although the ALP wentperilously close to defeat after one term in 2010. Of the ten PMs who have beenin office since World War II ended, (excluding the care-taker John McEwen), theaverage period of incumbency is a little over seven years.

What of post-prime ministerial longevity? At the time of writing 26 PMs haveleft office, beginning with Edmund Barton in 1903 and most recently KevinRudd in 2010. Three were ‘caretakers’ (Earle Page, Francis Forde and McEwen),who warmed the prime ministerial seat while a replacement was chosen for anincumbent who had died in office. They are excluded from these calculations.Six of the (remaining) 23 ex-PMs are alive: Gough Whitlam, Malcolm Fraser,Bob Hawke, Paul Keating, John Howard, and Rudd. Of the 17 that are deceasedthe average period between the final exit from office and death is 16 years.Consistent with Keane’s observation of a lengthening out of post-leadershiplives, the survival period for the living former PMs is trending higher.

Furthermore, the majority of the time between leaving office and the gravehas been spent outside parliament. Again eliminating the caretakers, formerPMs have remained in parliament following their final exit from office for anaverage four years and eight months. The indomitable William Morris Hughesis the outlier: his near three decades in the House of Representatives followinghis defeat easily exceeds the staying power of Scullin and Arthur Fadden, eachof whom remained for a further 17 years. Though surpassed by Fadden whowas treasurer for nine years in Robert Menzies’ post-World War II cabinets butwho is a special case because he was a ‘stop gap’ leader (Grattan, 2000: 9),Hughes’ record is also exceptional for the length and diversity of his ministerialservice following his premiership. While too compulsively ambitious and quarrelsome to settle into the mantle of elder statesmen, Hughes bears com-parison with Britain’s Arthur James Balfour; both ex-leaders became renownedfor their survival skills that included sliding between different party banners(Theakston, 2006: 450). Five other PMs have served in the ministry of a suc-cessor. The two examples since World War II are John Gorton and Rudd.Gorton’s decision to accept a portfolio from William McMahon was principallyactuated by a wrecking impulse, while it is uncertain yet whether Rudd will

210 ‘The Tideless Pond that Seemed Waiting for Me’

stay long in the ministry of his nemesis and successor, Julia Gillard. Moreover,the period between the loss of the prime ministership and the end of parlia-mentary career has shrunk in recent decades so that it is now usually mea-sured in weeks rather than in years.

The pre-modern era: Barton to Chifley

‘I feel like a school kid going on holidays’

So quipped James Scullin when he surrendered his commission in January1932 following his government’s crushing election defeat (Robertson, 1974:379). Scullin had greater cause than most to feel liberated at leaving office – his incumbency had been one long crisis as his administration helplesslygrappled with an economic maelstrom and was beset by internecine strife.Even so, Scullin’s insistence that, just as ‘he was not exhilarated when I was returned at the head of the Government’ he was ‘not downcast now’(Robertson, 1974: 378), rings true of the sentiment of Australia’s early PMs.It is striking how many evinced relief at the transition to life after office.Alfred Deakin, three times PM in the first Commonwealth decade, is anobvious case. Throughout large parts of his political career he was conflictedover whether to accept the obligations of public office or escape into a private world of spiritual and intellectual contemplation. The existentialhand-wringing was acute by his final premiership. When defeated at the1910 election, he professed to being ‘more relieved than anyone else I knowto find myself freer than I have felt for ten years past’ (La Nauze, 1979: 604).

There are other instances of early PMs not ‘raging’ against the dying ofthe political light. Barton set the precedent for graceful departure. Thoughinitially agonizing over whether he could ‘persuade myself to leave’, he‘showed little sign of regretting his separation from active politics’ oncesettled on the bench of the newly created High Court (Bolton, 2000: 297,310). Likewise, George Reid accepted with good humour being tapped onthe shoulder in 1908 to step aside as leader of the Anti-Socialist (formerlyFree Trade) Party as a necessary precondition for smoothing the way to thefusion of the non-Labor parties. Ensconced as Australia’s inaugural HighCommissioner in London by the time of the next election, Reid content-edly recorded: ‘To be out of the way is a strange and most pleasant experi-ence’ (McMinn, 1989: 257). Similarly, Joseph Cook ‘never gave the smallesthint that he coveted the leadership [and prime ministership]’ during late1916 and early 1917 when his Liberal Party was negotiating a coalitionwith the Hughes-led Labor breakaways (Murdoch, 1996: 115–27). Thenthere is Stanley Melbourne Bruce, who having also gone to England asHigh Commissioner following his government’s 1929 election loss, rejectednumerous overtures that he return to Canberra and reclaim the prime ministership (Cumpston, 1989: 106–11, 168).

Paul Strangio 211

Perhaps the early PMs protested too much and they were simply adept atdisguising their ambitions and relish of power. Hughes is a powerful coun-terpoint. Not only did he haunt the parliament for decades following hisdemise as PM, but his lust for office was never sated. As Donald Horne(2000: 182–3) cheekily observed of the octogenarian Hughes of 1945: ‘buteven if by now he had decided he might never again be prime minister – but how can one be certain? – he was still hoping for one more spell inoffice as attorney-general’. Yet, Hughes aside, the pattern is sufficientlystrong among the majority of the pre-Menzies PMs to suggest that thenotion of ex-leaders inevitably becoming hooked on office is informed byrecent rather than historical experience.

Why is that the PMs of yesteryear reconciled themselves more comfort-ably to handing over power? One supposition is that the prime ministerialproject was less individual in that era and thus there was not the samedegree of investment of self in the office. Perhaps also it is only since theprime ministership has become unambiguously central to the nation’s lifeand its powers so expansive that its seductions have become difficult torelinquish without separation angst. The physical toll that the office exactedon its occupants is another factor that influenced early PMs to yield to polit-ical mortality. Back then leaders had nothing like the support structuresavailable to their modern counterparts. Their resilience was also less in anera of comparatively poor diets, rudimentary medical treatments, higherrates of chronic morbidity and shorter life spans.

Poor health was an unwelcome reality for many of the early PMs; insome cases illness not only forced them out of the job but gravely com-promised their post-prime ministerial existence. Most severely, the burdensof office are regarded as having hastened the deaths of Lyons, Curtin andChifley. Although there were other extenuating factors, and he recoveredto live for another three and a half decades, ill-health contributed to thedecision of Labor’s first PM, John Christian Watson to resign the party leader-ship at the age of 40. Contemporary observers blamed his brief stint in officein 1904, which was said to have ‘knocked his nerves to tatters’ (McMullin,2004: 161). Similarly, a major reason for Barton relinquishing the primeministership in 1903 was that ‘doctors had advised him he would lengthenhis life expectancy by getting out of politics’ (Bolton, 2000: 294). For AndrewFisher and Deakin there would be no recovery. Both were suffering irrevers-ible health decline by their final premierships, beginning the descent intodementia that cruelly disabled and shortened their post-political lives (La Nauze, 1979: ch. 25–26; Day, 2008: 304, 328–9, ch. 13–14). Scullin also had his health ruined by the strain of governing during the depths of the Depression. Despite being in office for only two years, the physicalravages were manifest: ‘[Scullin] aged rapidly, his face becoming more lined,and his walk becoming slower … his hair changed from a sandy colour toabsolute whiteness’ (Robertson, 1974: 381).


Avenues of influence

In the first half of last century there were semi-established pathways opento ex-PMs that offered the potential for continuing influence. Staying inparliament and accepting further ministerial office was an option not onlyexercised by Hughes, but also by Cook and, briefly, by Bruce and Menzies(after his first premiership).

Scullin’s health was too impaired to enable him to serve in John Curtin’sWorld War II cabinet, but he was appointed to advisory positions. Scullin isan early example of the Labor Party elevating a former leader to the stationof elder statesman (Robertson, 1974: ch. 29). Watson’s influence over theLabor Party, by contrast, was cut short by his support for conscriptionduring World War I. He went on to play a role in the formation of the non-Labor Nationalist Party before drifting out of politics in the final decades ofhis life (Grassby and Ordonez, 2001: 136–41). A mesmerizing force over col-leagues during the early Commonwealth, the precipitous decay of Deakin’smind curtailed any de facto influence he might have retained following his departure from parliament. The post-prime ministerial authority ofDeakin’s predecessor, Barton, on the other hand, was undoubted as hetraded political leadership for judicial leadership in the final decade and a half of his life. From the High Court bench, Barton ruled on the consti-tutional integrity of his contemporaries’ legislation and presided in casesthat calibrated the balance of power between the newly created Common-wealth and states (Rutledge, 1979; Bolton, 2000: ch. 14). While his reputa-tion as a jurist is mixed, this places him in a unique post-prime ministerialcategory.

Whereas Barton’s elevation to the High Court proved a one-off, the HighCommissionership in London was a routine destination for the nation’s ex-leaders in the first half of last century. Four former PMs filled the postduring the three and a half decades from its 1909 inauguration: Reid(1909–16), Fisher (1916–21), Cook (1921–27) and Bruce (1933–45). While theoriginal expectation was that the High Commissioner would play an impor-tant diplomatic role, the post’s functions were largely ceremonial until the1920s. The chief retardant on the development of its diplomatic potential was‘the constraint imposed by prime ministers of the day’ (Bridge et al., 2010: 5;Meaney, 2010). At home, the post was popularly conceived as a plum for political has-beens. The Melbourne Herald grumbled in anticipation of Cook’s1920 appointment that it was ‘reserved for distinguished politicians who hadoutlived their usefulness in Australia, and may be adroitly consigned todignified ease’ (Attard, 1991: 11). For the ex-PMs who spent time as HighCommissioner, the position was of varying congeniality. ‘Fisher, still thinkingof himself in some ways a prime minister’ (Meaney, 2010: 47), had his tenureblighted by Hughes’ manic determination to sideline him at every turn andhis own failing health (Attard, 1995; Day, 2008: 361–400). By contrast, Cookand Reid better reconciled themselves to its limitations and found enough in

Paul Strangio 213

the role to provide a sense of purpose (McMinn, 1989: 253–69; Murdoch,1996: 137–9; Meaney, 2010).

It was Bruce who transformed the High Commissionership into a post of significance in the 1930s. Not only did he become an influential envoyin London, but also Australia’s lead representative at the League of Nationsin Geneva (including becoming chairman of its council). David Lee, authorof a forthcoming biography of Bruce, argues that this was a period when

Australia began for the first time to express its international personality.This was largely due to Bruce having become principal Australian dele-gate in Geneva … At a time when instructions from Canberra were irreg-ular, Bruce was left with considerable autonomy to shape Australia’sembryonic foreign policy (2010: 84).

The apogee of Bruce’s international diplomacy was the second half of the1930s: his counsel was sought on issues as varied as the Abyssinian crisis,the Royal abdication crisis and the Munich negotiations, and he oversawinitiatives within the League of Nations that prefigured the post-war develop-ment of the United Nations. As a proponent of appeasement, Bruce was dis-trusted by Churchill and his influence waned during World War II (Bridge,2010; Cumpston, 1989: ch. 8 and 9). Even so, Lee (2010: 82) judges him‘one of the most consequential Australian diplomats of the [20th] century’.Indeed, arguably, Bruce’s High Commissionership was of greater momentthan his premiership. That Bruce turned down several appeals in the 1930sto return to the Australian parliament suggests that he reckoned the oppor-tunities of international statesmanship from his base in London surpassedthose of antipodean domestic leadership.

Going ‘home’

In an era when London was the heart of the empire and Britain still ‘home’to many Australians, Bruce’s calculation was perhaps not so surprising. AsHorne (2000: 179) puts it: ‘London was the imperial city and Australia wasmerely a province; Bruce declined the offer of banishment’. Following theend of his High Commissionership, considering that there was ‘no nichefor me in Australia’, Bruce accepted a viscountcy and seat in the House ofLords. Though later appointed the inaugural chancellor of the AustralianNational University, he lived out his days in London (Radi, 1979).

Bruce was not alone among Australia’s early ex-leaders in settling in Britainfollowing his prime ministership, or in seeking recognition and achievementin the imperial metropolis. The syndrome was most pronounced among theformer non-Labor PMs, but not exclusively. Upon arriving in England in 1916to begin his High Commissionership, the Scottish-born Fisher declared thathe felt he had ‘returned not to a foreign country but to his home’. After brieflycoming back to Australia to settle his affairs following the end of his term,


Fisher returned to England. He even sought Labour Party nomination forthe House of Commons in the early 1920s, despite growing mental frailty(Day, 2008: 362, 401–8). Reid actually succeeded in gaining a seat in theCommons in early 1917 and was ‘mightily pleased’ by his (uncontested)election (McMinn, 1989: 271–2). Like Fisher, he died and was buried inLondon. Cook did return to Australia, but his biographer (Murdoch, 1996:139) suggests that for this former Staffordshire pit boy ‘the crowning delightof his life’ had been his acceptance ‘by the aristocrats of the Empire as oneof themselves’ during his High Commissionership.

Money matters

Part of the attraction of the High Commissionership was financial: it was arelatively lucrative post in an era when former PMs (and parliamentariansgenerally) received no pension. It was not until 1948 that the Chifley gov-ernment established a parliamentary pension scheme (Manthorpe, 2004).Accordingly, for at least the first half of last century, with the exception ofBruce who was independently wealthy, maintaining a source of incomewas an unavoidable concern for ex-leaders. This could be an incentive tostay in parliament as demonstrated by Scullin remaining in the House of Representatives long after he was an invalid. When he did leave in 1949, a testimonial was organized to keep him through his final years of life(Robertson, 1974: 478–9). Hughes was also the beneficiary of a testimonial,though in his case controversially while still PM. The handsome sum of£25,000, it had been ‘subscribed in recognition of his services to Australiaand the Empire during the war and at the [Versailles] Peace Conference’(Fitzhardinge, 1979: 456–7). Reid’s decision to take up a seat in the Houseof Commons as a septuagenarian certainly had a pecuniary motive. Eventhen he relied upon extra income from directorship fees to save him from‘financial embarrassment’ during his autumn years (McMinn, 1989: 273).Similarly, Fisher’s biographer surmises that ‘financial reasons’ induced himto bear the indignity of seeking endorsement for the House of Commonswhen already shadowed by dementia (Day, 2008: 407–8). Barton also remainedon the High Court well after his health had seriously declined motivated inpart to provide for his family (Bolton, 2000: 328).

Other former PMs found alternative means to the public purse to finan-cially sustain them. Perhaps the most remarkable story is Watson’s. ‘[L]iter-ally broke’ and ‘desperate for an income to survive’ (Grassby and Ordonez,2001: 119), when he left parliament in 1910 he embarked on variousmoney-making ventures. These were initially unsuccessful: he returnedempty-handed from a gold prospecting mission to South Africa and landspeculation in outer Sydney also failed. Yet Watson did not abandon hisbusiness ambitions. He became a director of a major wool and textile com-pany and pioneering lobbyist on its behalf. In the 1920s, the expandingmotoring industry became a focus of his activities. He was a founder and

Paul Strangio 215

driving force of the National Roads and Motorists’ Association, and chair-man of the New South Wales Traffic Advisory Committee. In subsequentyears, he was the first chairman of directors of the Australian MotoristsPetrol Co. Ltd (AMPOL) and a founding director of Yellow Cabs AustraliaLtd (Grassby and Ordonez, 2001: ch. 10–12). While his estate was relativelymodest when he died in 1941, Watson’s extensive business interests give lieto the notion that it is only in recent times that former Labor leaders havesoiled their hands with commercial enterprises.

‘I begin to regret that I ever retired’: The Menzies transition

The political afterlife of Sir Robert Menzies may be viewed as a hinge betweentwo eras. Most strikingly, Menzies post-prime ministerial disenchantmentwith his political heirs heralded a trend to which few of his successors have been immune. As the only unambiguous example of an Australian PM who relinquished office voluntarily, Menzies was perhaps an unlikelyvictim of that syndrome. His resignation in 1966 at age 71 came after anunparalleled seven consecutive election victories and unbroken 16 years in office since 1949. He had nothing left to prove. On the other hand, his lengthy supremacy of the Australian political landscape and mantle as the father of the modern Liberal Party may have been the very things that made it difficult for him to tolerate lesser men steering the ship of state.

Disillusionment surfaced quickly. Already uncomfortable with the newbroom attitudes of his successor Harold Holt (Henderson, 1994: 184), Menziesdisapproval of the Liberal Party’s direction spiked after John Gorton replacedHolt as PM following the latter’s drowning at the end of 1967. Gorton’s first major sin was to overlook Menzies’ long-time bureaucratic right-handman, John Bunting, when appointing a new head of the Prime Minister’sDepartment. Menzies fumed to another senior official: ‘I begin to regret that I ever retired!’ Gorton further provoked Menzies’ wrath by cancelling entitle-ments that he had previously enjoyed while carrying out his duties as Warden of the Cinque Ports in England. He bitterly complained that Gortonhad decided to ‘suppress all reference to, or traces of, Menzies’ (Martin, 1999:557).

Menzies’ most accomplished biographer, A. W. Martin (1999: 558), con-jectured these ‘extreme’ responses were a ‘delayed reaction at having vol-untarily given up the mana he had so long taken for granted’. If so, thatreaction intensified in the final decade of his life. Privately, he renouncedthe Liberal Party. His daughter, Heather, recalled her father bridling whenan interlocutor referred to it as ‘your party’. ‘My party? It’s not my party’,he indignantly retorted (Henderson, 1994: 183). Indeed, it seems likely thatthe father of the Liberal Party committed the ultimate act of filial disown-ment by spurning his political child at the ballot box at the 1969 federal


election. Gorton’s removal as Liberal leader failed to restore his faith; he repeated the snub in 1972 and, possibly, in 1974 (Henderson, 1994:186–7; Martin, 1999: 564–5). By then he had succumbed to deep pess-imism about the state of Australian politics. The Whitlam Labor gov-ernment appalled him and, while welcoming Malcolm Fraser’s advent as Liberal leader and then PM, Fraser’s early years in office did not measure up to his expectations. Menzies took to advising Fraser ‘on every-thing from election tactics to the improvement of his television appear-ances’ (Martin, 1999: 546, 562–5). If this meddling portended a futuretrend of ex-PMs peering over the shoulders of successors, Menzies at least kept his interventions behind-the-scenes rather than publicly airingthem.

Martin has speculated that stroke-induced personality changes were acause of Menzies’ proclivity for hyper-critical political and personal judge-ments in his final years of life (Martin, 1999: 562). The rather precipitousdecline of Menzies’s health at the conclusion of his political career is, as we have seen, a misfortune he shared with many of his predecessors. Thiswas as much as anything a corollary of the advanced age at which heexited office. While still in reasonable health when he resigned, Menziessuffered a stroke in late 1968. He recovered slowly, but further strokes in 1971 permanently weakened him and curtailed his activities until hisdeath in 1978 (Martin, 1999: 558–61). In another way, Menzies was morefortunate than his forerunners in that he benefited from the introductionof parliamentary pensions. As a result of revisions to that scheme in the1950s, ex-PMs had become entitled to additional allowances to those givento other retired MPs. Menzies was also the first former PM to be grantednon-pension entitlements. However, Menzies was not wealthy when heretired. A group of friends and admirers ‘clubbed together’ to purchase a home in the Melbourne suburb of Malvern into which he and his wifePattie shifted after vacating the Lodge (Martin, 1999: 529, 546, 561;Manthorpe, 2004).

Menzies’ publication of memoirs in retirement (1967 and 1970) also anticipated future directions. Few of his predecessors had done so. Reid had been the first to produce a memoir after leaving office, whileHughes penned two ‘collections of anecdotes’ (Reid, 1917; Hughes, 1947,1950; Horne, 2000: 185). Otherwise, and perhaps incongruously, it was the short-lived Country Party PMs, Page and Fadden, who completedmemoirs (Page, 1963; Fadden, 1969). Other early PMs curtly dismissed the idea (Cumpston, 1989: 269; Murdoch, 1996: 140). In common withnearly all those produced by his prime ministerial counterparts before and since, Menzies’ memoirs lacked insight and were personally unreveal-ing. One of the few interesting sections in the first volume were his reflec-tions on the nation’s other leaders. Disenchanted with his successors, it wasironic that Menzies accused Hughes of becoming ‘primarily reminiscent’

Paul Strangio 217

in later life. Without hint of self-consciousness, he quoted Shakespeare’s Henry V:

Old men forget; yet all shall be forgot,But he’ll remember with advantagesWhat feats he did that day … (Menzies, 1967: 109).

A final way in which Menzies’ post-prime ministerial activities were tran-sitional was his enthusiastic, but archaic, answer to the call of empire.Before retirement Menzies had been awarded the Order of Thistle, thehighest order of knighthood yet conferred on an Australian, and in mid-1966 was installed as Constable of Dover Castle and Warden of the CinquePorts in an elaborate traditional ceremony. He carried out the latter office conscientiously as long as his health permitted, making annual pil-grimages to Walmer Castle in Kent, the official residence of the Warden(Joske, 1978: 347–53). The images of Menzies bedecked in Admiralty uni-form and revelling in the pageantry of the position was fodder for critics.By then the sun was well and truly setting on the British empire and the economic and the cultural ties between Australia and United Kingdomwere thinning, making it fashionable to mock Menzies’ Anglophile devotions (e.g. Horne, 1964: 97–102, 201–13). However unfair were the caricatures of him as an antediluvian and obsequious empire loyalist, those reactions underscored that the days of ex-PMs finding solace in the bosom of the mother country were past. Both Gorton and McMahonaccepted imperial knighthoods in the future (Hancock, 2002: 382), butincreasingly Australia’s former leaders had to find international recognitionelsewhere.

The modern era: Gorton to Rudd

‘The body keeps twitching after the head is cut off’

In his memoir of the Keating prime ministership, Don Watson, Keating’sspeechwriter, describes the aftermath of the government’s defeat by theHoward-led Liberal-National Coalition in 1996:

for months after the loss we would act as if we still had it in our powerto rectify things. I kept a file of Howard’s broken promises; composed inmy head speeches that Kim Beazley3 – or Paul Keating – might deliver;sent the odd note to Beazley’s office; for nearly three months kept thepolitical diary going … Paul Keating, while maintaining an admirabledignity and despite his depression, felt each blow on his good name as itis said people continue to feel a limb after it has been amputated. I felt it


too … Political death is like the other kind – the body keeps twitchingafter the head is cut off (Watson, 2002: 732).

Watson writes with an artist’s licence. Nonetheless, his evocation of a senseof bereavement and disorientation at the loss of office is in tune withmodern experience. In the place of the resigned relief of their earlier coun-terparts, ‘raging’ against political death seems to have been far more thenorm among the modern ex-PMs. This has materialized in a pronouncedtendency for former office-holders to insinuate themselves into publicdebate and look over the shoulder of leadership successors. The relation-ship between ex-PMs and their successors has frequently been strained and,in a pattern notable on the Liberal side of politics, sometimes poisonous.No matter the partisan stripe, modern former PMs have also consistentlyindulged in self-justification through defence and advocacy of their primeministerial record.

Why is it that the transition to life after office has been more problematicfor former leaders in recent decades? One answer is that this is a mediacreated phenomenon. In early times, ex-PMs could, if they desired, slideinto relative obscurity. This is no longer the case. As Keane points out(2009: 290), the pervasiveness of the media means that it is now ‘virtuallyimpossible’ for former heads of government to ‘stay off stage, or to remaininvisibly silent’.

The case of Howard (PM from 1996–2007) is instructive. Both before andsince leaving parliament he insisted on his determination to eschew com-menting upon politics in retirement (e.g. van Onselen, 2009). In doing so,he was implicitly critical of other ex-PMs for sullying themselves by gettingcaught up in the news media cycle. Likewise, following his defeat, Howardadmirers predicted that he would be publicly circumspect compared to hisoutspoken predecessors (e.g. Switzer, 2008). Yet, since his removal fromoffice, chased by journalists, Howard has ventilated his opinions aboutpolitical developments with growing frequency. This has included com-menting on leadership in the Liberal Party (e.g. Australian BroadcastingCorporation, 2010a).

Secondly, it is plausible that modern era ex-PMs have become moresusceptible to separation angst because of the evolution of the office. Incommon with other comparable democracies (Poguntke and Webb, 2005),since the 1970s there has been an augmentation of leadership resourcesand increase in the prominence of office-holders: the prime ministership ismore than ever before the axis around which the nation’s political systemturns (Walter and Strangio, 2007). The trappings of office have also beenenormously enhanced: contrast, for example, the grandeur and vast scaleof the executive suites of the new parliament house (opened in 1988) withthe modesty of the PM’s office in the old parliament. While the prime min-istership might have been a poor aphrodisiac in bygone days, its powers of

Paul Strangio 219

seduction and addiction have been supercharged in modern times. Weaningoneself off the job has undoubtedly become a greater practical and psycho-logical challenge. Malcolm Fraser’s wife, Tamie, has noted that the firstyears out of office were extremely difficult for her husband, who was ‘morelost than I think I realised at the time’. He even had to (re-)learn ‘how todial out a number from home … and book a plane’ (Fraser and Simons,2010: 620).

Howard’s insistence that he has smoothly acclimatized to life after theprime ministership was belied by a continuing preoccupation with politics.He maintained an office that bears an uncanny resemblance to his primeministerial suite complete with cable television tuned so ‘he can stay abreastof political developments as they happen’ (van Onselen, 2009). It was notuntil 2010, when he was nominated (unsuccessfully) for high office in theInternational Cricket Council, that he seemed to have settled on a lifeoutside Australian party politics.

In a similar vein, the diaries of Mark Latham, Labor opposition leader2003–05, refer to the ‘separation anxiety’ suffered by two former Labor PMsWhitlam and Keating. Latham, a Whitlam protégé, records an encounterwith a moody Keating in 2002:

He’s still obsessed with the Federal Parliament … Reminds me of myworst feelings about Gough [Whitlam] – they can’t let go, politics as anobsession, instead of an achievement in life from which you move on(Latham, 2005: 187, 407).

Third, is the constitutional resilience of modern ex-PMs. Notwithstanding ageneral trend towards a ‘younging’ of parliamentary politics (Weller andFraser, 1987), the post-Menzies PMs have on average been a little olderthan their forerunners when leaving office. But they have been able toweather its strains without serious or irreversible damage to their physicalwell-being. To take one example: when Fraser was defeated in 1983 at theage of 52 he has been described as still being in his ‘prime’ (Fraser andSimons, 2010: 612). Yet Deakin, Fisher and Scullin all departed office at asimilar age and were in serious physical decline. With their health intact,fallen leaders are nowadays much more equipped to forge a post-politicalpublic life – Fraser notes he was not at an age when he was ‘ready to retire’(Fraser and Simons, 2010: 612). But the extra vitality might also rendermodern ex-leaders more prone to feelings of premature obsolescence andfantasies of their capability for continuing command.

‘Raging’ against successors

Nearly all of the modern former PMs have been unable to resist sporadic-ally sniping at successors through the media. The efficacy of such interven-tions has generally been limited even when focused on substantive policy


questions. Criticisms of party opposites are usually treated as part of thepartisan bi-play of politics and quickly shrugged off. They can also rebound.Keating’s frequent (and choleric) attacks on Howard during the latter’s pre-miership were seen as reflecting badly on him and the Labor opposition.Most arresting, however, have been relations on the Liberal side, whichhave been characterized by spectacular fallings-out between ex-officer holdersand the party. Gorton’s wrath towards his Liberal successors was nourishedby his defeat as PM in a party room coup in March 1971. A catalyst hadbeen Fraser’s resignation from his ministry, who departed alleging disloy-alty by Gorton. His other nemesis was McMahon, his deputy and the manwhose ambition for the prime ministership had been thwarted whenGorton replaced Holt in 1968. When McMahon deposed Gorton in 1971,Gorton improbably accepted the deputy leadership but was soon dismissedfor destabilizing conduct. He then used the freedom of the backbench toroutinely dissociate himself from Liberal government policy. Gorton returnedto the Liberal front bench after McMahon lost office, only to resign in furywhen Fraser became leader in 1975. The breach between Gorton and LiberalParty was completed when he stood as an independent for the Senate at anelection later that year and advocated a vote for Labor in the lower house(Hancock, 2002: 337, 346–62, 376–80).

Fraser’s alienation from the Liberal Party during Howard’s prime minister-ship was more obviously anchored in philosophical differences. He emergedas an insistent public critic of the government’s position on human rights andits foreign policy. More generally, Fraser deplored the party’s realignmenttowards the conservative end of the ideological spectrum and diminution of ‘liberal principles’ under Howard’s leadership (Stewart, 2005; Fraser andSimons, 2010: ch. 22). So complete was Fraser’s disillusionment with theHoward government by its latter stages there are strong hints that, like Men-zies before him, he disowned the Liberal Party at the ballot box (AustralianBroadcasting Corporation, 2010b).

For former leaders, the results of these long-standing feuds have been atbest mixed. Fraser’s criticisms of Howard ‘brought him more media atten-tion than he had had for years’ and gained him new-found admirers amongthose ‘who had once reviled him’ (Fraser and Simons, 2010: 734–5). Oncedespised by the left, by the 2000s Fraser was receiving standing ovations inpacked university auditoriums. On the other hand, as Kevin Theakston hasobserved (2006: 449) in respect to the development of similarly antagonisticrelations within the Conservative Party in Britain, such conduct can also be ‘self-defeating’ for ex-leaders by ‘costing’ them influence over their ownparty’. Gorton and Fraser both suffered this fate. Gorton was ostracized by theLiberals for many years. Only in the last years of his life was he rehabilitated(Henderson, 2000: 300–1). Equally, Fraser’s regular sallies against Howard lefthim isolated in the Liberal Party that he hoped to influence. He had to endurecountless humiliations during the Howard era. For example, at a 2001 Liberal

Paul Strangio 221

Party conference parts of the audience consciously snubbed Fraser by clattering cutlery as he was presented with a tribute on stage. Young Liberalorganizations expressed their hostility to him by demanding he resign fromthe party (Hooper, 2005; Stewart, 2005): a step he seriously contemplated(Fraser and Simons, 2010: 714). This can be contrasted to the Labor side, where the tradition has been to treat ex-PMs with polite deference,even when their public pronouncements, viz. Keating, try the patience of successors and the party.

‘Self-justification’

Asked in 2005 why he had not written memoirs, Fraser responded that auto-biographies too frequently became an exercise in ‘self-justification’ (Stewart,2005). Though not the only outlet for defending the prime ministerialrecord, memoirs have become virtual standard practice for former leaderssince the era of Menzies. Howard has dedicated much of his time since his 2007 defeat to chronicling his political career (van Onselen, 2009),while Fraser has recently completed a major autobiography. Conscious of his previous disavowal of the genre, Fraser insisted that his purpose was to address ‘a new generation about the true nature of liberalism’ (Fraser and Simons, 2010: 736). Their Liberal predecessors have also putpen to paper in defence of their prime ministerships. Gorton wrote a series of articles for Rupert Murdoch’s Sunday Australian vaingloriouslytitled ‘I Did It My Way’ (Hancock, 2002: 346–50). McMahon, usuallyjudged the most undistinguished of Australia’s prime ministers in any era, hawked his memoirs to publishers but without success (Abjorensen,2007).

It is Whitlam, however, who set the benchmark for post-prime ministerialself-justification through print. He began with a closely argued account ofthe 1975 constitutional crisis (1979), which was later followed by an exhaus-tively researched and painstakingly comprehensive magnum opus on thepolicy-making record of his government (1985). He clearly regarded thelatter as not merely a vehicle for setting the record straight on his primeministership, but also as a catechism for the Labor Party going forward.Hawke’s memoir (1994) was also a door-stopper. The reaction to it illus-trated that self-championing can be a counter productive exercise for formerPMs. The book was roundly criticized as an exercise in self-aggrandizementand taken as confirmation that its author was a narcissist extraordinaire. He was also attacked for vindictiveness to Keating, the man who had prised the prime ministership from Hawke in 1991 (e.g. Anson, 1994;Carlyon, 1994; Jones, 1994). Ironically, it is Keating, who, despite eschewingmemoirs, has probably been most unrelenting of recent ex-office holders inextolling his executive legacy. This has been complemented by his equalpreoccupation with tearing down the record of Howard who beat him at the1996 election.


Moving on and filling the void

What have former PMs of the modern era done in retirement and how has itdiffered from their predecessors? One thing they have not done is linger in par-liament. McMahon is the exception, spending nine forgettable years on thebackbench following his government’s defeat and towards the end irritatingthe Liberal Party with his reluctance to vacate his seat (Henderson, 1997).Whitlam is unique among the modern PMs in that he resumed as oppositionleader after losing office in 1975. It was an unhappy experience that endedignominiously when he led Labor to a second successive landslide defeat in1977 (Ramsey, 2001: 122–5). Thereafter fallen PMs resigned the party leader-ship and exited parliament with almost indecent haste. After losing the 1983election, Fraser told one former colleague that he feared if he had not quit ‘theparty would have torn itself apart gnawing over my bones’. Ironically, givenwhat was to come, he explained to another supporter that if he stayed on thebackbench he ‘could so easily become a source of difficulty and division’(Fraser and Simons, 2010: 614–15). Hawke and Keating both followed Fraser’sexample by leaving the parliament within weeks of being deposed. The issuedid not arise for Howard because he lost his seat when his government fell in2007, but by then the pattern had become so established that it seemed almostunthinkable that any ex-PM would seek to remain in parliament let alone haveanother tilt at office. Rudd, however, has defied that trend, although only timewill tell how meaningful and durable will be his post-leadership parliamentarycareer. Notably, this development of ex-leaders promptly leaving parliament isalso apparent in the sub-national sphere of Australian politics as well as inother democracies such as Britain. Keane welcomes it for the reason that heargues it draws a clear demarcation between ‘holding and not holding office’and militates against the recycling of political elites that occurs in a countrylike Italy (2009: 286–8). In a practical sense, it is another factor extending thelength of political afterlives as well as compelling re-invention.

In another way, modern ex-PMs are treated as anything but dispensable.In addition to a handsome parliamentary pension, they also receive gen-erous additional entitlements that preserve their special status. Those enti-tlements have been progressively augmented. When Howard entered thepost-political world in 2007 his retirement perks included a prime min-isterial gold travel pass allowing him and his wife to up to 40 domestic airflights per year, two staff, an office, car and driver (Hart, 2008). Arguably,these trappings blur the line to which Keane refers as Howard’s retirementfaux prime ministerial office possibly attests.

With the option of remaining in parliament receding and with no compar-able pathway as the High Commissionership, modern ex-PMs have occupiedtheir political retirements with a diverse range of activities.4 Despite generouspensions, few have been slow to pursue commercial and business activities.They have been able to do so in the absence of regulatory constraint. Aus-tralia lagged behind comparable democracies in adopting codes of conduct in

Paul Strangio 223

relation to post-parliamentary employment of ministers and only recentlyimposed a brief cooling off period on lobbying activities (McKeown, 2009). Insome cases, former PMs have abided by self-imposed limits. Fraser abortedconsultancy activities when he discovered that ‘companies wanted not hisadvice but for him to use his contacts and connections on their behalf’ (Fraserand Simons, 2010: 612). Other ex-office holders have been less circumspect;for example, McMahon was an advisor to various financial institutions,including the Bank of America. Publications advances for memoirs and lecturefees have been other staple earners for former leaders. Whitlam took things inquite another direction by cashing in on his larger than life reputation byappearing in advertisements endorsing photocopier and pasta sauce products.It is Hawke, though, who has become most notorious for hanging out hisshingle once he left parliament as he built international business interestsfrom his Sydney base (Brown, 2002: 182, 201). Yet his is also a cautionary talefor former leaders because his enthusiasm for money-making activities, com-bined with a well-publicized upheaval in his private life provoked censoriouscriticism (e.g. Blewett, 2000: 405).

However, ex-PMs of the modern era have not merely occupied their polit-ical afterlives chasing financial enrichment. They have also busied themselvesin civil society. Whitlam’s activities have been prolific. To name a few: he wasa visiting fellow at the Australian National University in 1978 and visitingprofessor at the University of Adelaide in 1983, a mem-ber of the Consti-tutional Commission in the mid-1980s, chairman of the Australia-ChinaCouncil (1986–91), chairman of the National Gallery of Australia (1987–90),and a member of the successful 1995 Sydney Olympic bid team. Fraser’s mostprominent post-prime ministerial role was his founding in 1987 and chair-manship for the next decade and a half of CARE Australia, which was part ofthe network of the international CARE humanitarian aid organizations.Hawke co-chaired a 2002 review of the ALP following its third consecutivefederal election loss the year before. He has also been an active patron of the Bob Hawke Prime Ministerial Centre and Hawke Research Institute, both based at the University of South Australia. Keating has been a visitingprofessor of public policy at the University of New South Wales.

Nor have post-prime ministerial activities been confined to the domesticsphere. Coming from a geographically distant middle power, Australia’s formernational leaders are less advantageously situated than, say their Europeancounterparts, to become involved in the proliferating range of ‘governmentaland non-governmental structures that operate at the regional and global level’(Keane, 2009: 293–5). Moreover, the transition to the international arena hasrelied to a significant extent upon sponsorship by successors, which not all ex-holders have enjoyed. The Hawke government has been most generous. Itappointed Whitlam Australian Ambassador to the Paris-based United NationsEducational, Scientific and Cultural Organization (UNESCO). Whitlam alsoserved on the Independent Commission on International Humanitarian


Issues and the World Heritage Committee while with UNESCO, and laterchaired the General Assembly of the World Heritage Convention.

Fraser is the other ex-PM to have strutted his stuff on the internationalstage. He was nominated by the Hawke government as Australia’s represen-tative on the Commonwealth Group of Eminent Persons against Apartheidin South Africa, a grouping he co-chaired. He was also chair of the UnitedNations Secretary’s Expert Group on African Commodity Issues. His bid tobecome secretary-general of the Commonwealth failed, however, despiteAustralian government support. During the 1990s, Fraser was president andvice president of CARE international.

Notwithstanding these achievements, neither Whitlam nor Fraser hasmatched the post-office international eminence attained by Bruce. Theirsojourns also had the whiff of exile. Fraser consciously pursued inter-national activities to remove himself from the domestic political scene(Fraser and Simons, 2010: 621) – a calculation that no doubt also occurredto his (and Whitlam’s) patrons in Canberra. Rudd’s angling for, andappointment by Julia Gillard to, the foreign affairs portfolio may in parthave had a similar rationale.

Conclusions

Prime ministerial afterlives have changed markedly in Australia. In the firsthalf of the last century, it was not unusual for leaders to leave office feelingliberated, fagged out and impecunious. For much of that era, the HighCommissionship in London served as a way station in the transition frompublic life for former PMs. It afforded financial sustenance and compen-satory prestige for the loss of executive office, and in the case of Brucebecame a platform for a major post-office career of international diplo-macy. The High Commissionership also transported occupants outside the orbit of day-to-day Australian political life. In other cases, ill-health orpremature death cut short or inhibited post-prime ministerial lives andremoved ex-leaders from public consciousness.

By contrast, the pattern that has congealed over the past four decades is for former PMs to be still vital upon leaving office, materially secure, rest-lessly active and regularly conspicuous. Several reasons have been proposedfor this transformation. The increased centrality of the prime minister-ship to the nation’s political life and growth of its perks have undoubtedlyrendered it more addictive for incumbents. Moreover, extended longevityhas increased the prospect of there being alive at any one time a cohort of ex-PMs of able body and mind, while their access to generous pen-sions and other entitlements ensures they are also well resourced. Finally,the media fishbowl of modern life has greatly diminished the likelihood of prime ministerial survivors quietly slipping into obscurity. The grow-ing ubiquity of ex-PMs has coincided with an era in which previously

Paul Strangio 225

established post-office paths have eroded. It has become unorthodox fordeposed leaders to stay in parliament and seek further ministerial office,and the option of the High Commissionership is also long gone. Newopportunities have opened up, particularly in the international sphere, butthe modern trends of post-prime ministerial public life remain on thewhole inchoate.

Public life after office has thus become both more commonplace andvariegated. The challenges have changed. Former PMs no longer have to worry much about economic sustenance, but they wrestle with otherinsecurities attendant with heightened office dependency and inventing a meaningful (and prolonged) life in retirement. They have become morepublicly fractious: peering over the shoulders of successors, defending pastrecords and demanding attention in other ways.

We should not confuse noise with influence. Australia passes Keane’s test for a well-defined demarcation between holding and leaving office. Not only do ex-leaders now get out of parliament hastily, but their retire-ment activities raise few concerns about them being recycled into gov-ernment or linked power networks. Labor has been a better track record of elevating its former leaders to elder statesmen, but what influence they retain is more symbolic than substance: parties move on. What also seems to hedge the influence of ex-leaders in Australia is an ingrainedcultural apathy, if not intolerance, towards them. In a country that has never had a strong tradition of memorializing or venerating formerleaders, they are still largely ignored and their interventions reflexivelygiven short shrift. There is little allowance for the ‘good works’, which ex-office-holders can carry out as guardians of good public policy andethical standards in executive governance and as virtuous agents in civilsociety (Keane, 2009: 295). While he was still in office Australia’s mostrecently retired PM, Howard, mused about a ‘more mature approach’ to the nation’s ex-leaders (Henderson, 1997). It is something the country stillawaits.

Notes

1 G. H. Reid (1917) My Reminisces (London: Cassell and Company), p. 365.2 For much of the basic empirical data underpinning this chapter, I have relied

on B. Galligan and W. Roberts (eds) (2007) The Oxford Companion to AustralianPolitics (Melbourne: Oxford University Press), pp. 655–7; and the NationalArchives of Australia, ‘Australia’s Prime Ministers’ website. Accessed March 2010http://primeministers.naa.gov.au/primeministers/.

3 Keating’s successor as Labor Party leader.4 Much of this information comes from the National Archives of Australia, ‘Australia’s

Prime Ministers’ website. Accessed March 2010 http://primeministers.naa.gov.au/primeministers/.


Bibliography

Abjorensen, N. (2007) ‘Picture incomplete’, National Library of Australia News. AccessedNovember 2008 http://www.nla.gov.au/pub/nlanews/2007/jan07/story-3.pdf.

Anson, S. (1994) ‘The Hawked memoirs’, Age, 20 August.Attard, B. (1991) ‘The Australian High Commissioners’, Working Papers in Australian

Studies, 68, Sir Robert Menzies Centre for Australian Studies, Institute of Common-wealth Studies, University of London, London.

Attard, B. (1995) ‘Andrew Fisher, the High Commissionership and the collapse ofLabor’, Labour History, 68, 115–31.

Australian Broadcasting Corporation (2010a) ‘The authentic Mr Abbott’, Four Corners,19 March. Transcript accessed March 2010, http://www.abc.net.au/4corners/content/2010/s2846485.htm.

Australian Broadcasting Corporation (2010b) ‘Malcolm Fraser produces his memoirs’,The 7.30 Report, 23 February. Transcript accessed March 2010, http://www.abc.net.au/7.30/content/2010/s2828288.htm.

Blewett, N. (2000) ‘Robert James Lee Hawke’, in M. Grattan (ed.) Australian PrimeMinisters (Sydney: New Holland).

Bolton, G. (2000) Edmund Barton: The One Man for the Job (Sydney: Allen and Unwin).Bridge, C. (2010) ‘“Undependable busybody”? S.M. Bruce and World War II’, in

C. Bridge, F. Bongiorno and D. Lee (eds) The High Commissioners: Australia’sRepresentatives in the United Kingdom, 1910–2010 (Canberra: Department of ForeignAffairs and Trade).

Bridge, C., Bongiorno, F. and Lee, D. (2010) ‘Introduction’, in C. Bridge, F. Bongiornoand D. Lee (eds) The High Commissioners: Australia’s Representatives in the UnitedKingdom, 1910–2010 (Canberra: Department of Foreign Affairs and Trade).

Brown, W. (2002) Ten Prime Ministers: Life among the Politicians (Sydney: LonguevilleBooks).

Carlyon, L. (1994) ‘Hawke’s memoirs are nothing if not a fine romance’, Age, 18 August.

Cumpston, I. M. (1989) Lord Bruce of Melbourne (Melbourne: Longman Cheshire).Day, D. (2008) Andrew Fisher: Prime Minister of Australia (London: Fourth Estate).Fadden, A. W. (1969) They Called Me Artie: The Memoirs of Sir Arthur Fadden (Brisbane:

Jacaranda).Fitzhardinge, L. F. (1979) The Little Digger, 1914–1952: William Morris Hughes A

Political Biography Volume II (Sydney: Angus & Robertson).Fraser, M. and Simons, M. (2010) Malcolm Fraser: The Political Memoirs (Melbourne:

Miegunyah Press).Gold, A. (2007) ‘Yesterday’s men should keep silent’, Australian, 11 June.Grassby, A. and Ordonez, S. (2001) John Watson (Melbourne: Black Inc.).Grattan, M. (ed.) (2000) Australian Prime Ministers (Sydney: New Holland).Hancock, I. (2002) John Gorton: He Did It His Way (Sydney: Hodder).Hart, C. (2008) ‘Retiree Howard advertising for a helper’, Australian, 7 January.Hawke, B. (1994) The Hawke Memoirs (Melbourne: William Heinemann).Henderson, G. (1994) Menzies’ Child: The Liberal Party of Australia, 1944–1994 (Sydney:

Allen & Unwin).Henderson, G. (1997) ‘The fate of former PMs’, Sydney Morning Herald, 13 May.Henderson, G. (2000) ‘Sir John Grey Gorton’, in M. Grattan (ed.) Australian Prime

Ministers (Sydney: New Holland), pp. 299–311.Hooper, C. (2005) ‘Young Liberals in the chocolate factory’, The Monthly, 2, 19–27.

Paul Strangio 227

Horne, D. (1964) The Lucky Country: Australia in the Sixties (Melbourne: Penguin).Horne, D. (2000) Billy Hughes (Melbourne: Black Inc).Hughes, W. M. (1947) Crusts and Crusades: Tales of Bygone Days (Sydney: Angus &

Robertson). Hughes, W. M. (1950) Policies and Potentates (Sydney: Angus & Robertson).Jones, B. (1994) ‘Unremarkable memories’, Age, 17 August.Joske, P. E. (1978) Sir Robert Menzies, 1894–1978: A New, Informal Memoir (Sydney:

Angus & Robertson).Keane, J. (2009) ‘Life after political death: The fate of leaders after leaving high

office’, in J. Kane, H. Patapan and P. ’t Hart (eds) Dispersed Democratic Leadership:Origins, Dynamics, and Implications (Oxford: Oxford University Press), pp. 279–98.

La Nauze, J. A. (1979) Alfred Deakin: A Biography (Sydney: Angus & Robertson).Latham, M. (2005) The Latham Diaries (Melbourne: Melbourne University Press).Lee, D. (2010) ‘“Ambassador-at-large par excellence”: S.M. Bruce and the League of

Nations’, in C. Bridge, F. Bongiorno and D. Lee (eds) The High Commissioners:Australia’s Representatives in the United Kingdom, 1910–2010 (Canberra: Departmentof Foreign Affairs and Trade).

McKeown, D. (2009) ‘A survey of codes of conduct in Australian and selected overseasparliaments’, Background Note, Department of Parliamentary Services, Parliament ofAustralia. Accessed March 2010 http://www.aph.gov.au/Library/pubs/BN/pol/Codes-OfConduct.pdf.

McMinn, W. G. (1989) George Reid (Melbourne: Melbourne University Press).McMullin, R. (2004) So Monstrous a Travesty: Chris Watson and the World’s First National

Labour Government (Melbourne: Scribe Publications).Manthorpe, L. (2004) ‘The Parliamentary Retiring Allowance Act 1948: Debates,

committee reports, remuneration tribunal reviews and a chronology of legislativeamendments’, E-Brief, Parliamentary Library, Parliament of Australia. AccessedNovember 2008 http://www.aph.gov.au/library/intguide/POL/praa.htm.

Martin, A. W. (1999) Robert Menzies: A Life, Vol. 2, 1944–1978 (Melbourne: MelbourneUniversity Press).

Meaney, N. (2010) ‘The first high commissioners: George Reid and Andrew Fisher’, in C. Bridge, F. Bongiorno and D. Lee (eds) The High Commissioners: Australia’s Repres-entatives in the United Kingdom, 1910–2010 (Canberra: Department of Foreign Affairsand Trade).

Menzies, R. G. (1967) Afternoon Light: Some Memories of Men and Events (Melbourne:Cassell Australia).

Menzies, R. G. (1970) The Measure of the Years: Prime Minister of Australia, 1939–41and 1949–66 (Melbourne: Cassell Australia).

Murdoch, J. (1996) Sir Joe: A Political Biography of Sir Joseph Cook (London: MinervaPress).

Norton Smith, R. and Walch, T. (eds) (1990) Farewell to the Chief: Former Presidents inAmerican Public Life (Wyoming: High Plains Publishing).

Page, E. (1963) Truant Surgeon: The Inside Story of Forty Years of Australian Political Life(Sydney: Angus & Robertson).

Poguntke, T. and Webb, P. (2005) The Presidentialization of Politics: A ComparativeStudy of Modern Democracies (Oxford: Oxford University Press).

Radi, H. (1979) ‘Bruce, Stanley Melbourne [Viscount Bruce] (1883–1967)’, inAustralian Dictionary of Biography, vol. 7 (Melbourne: Melbourne University Press),pp. 453–61.

Ramsey, A. (2001) ‘The Hayden years’, in J. Faulkner and S. Macintyre (eds) TrueBelievers: The Story of the Federal Parliamentary Labor Party (Sydney: Allen & Unwin).


Ray, R. (2008) ‘Desperately seeking leadership’, Herald-Sun, 20 September.Reid, G. H. (1917) My Reminisces (London: Cassell and Company).Robertson, J. (1974) J. H. Scullin: A Political Biography (Perth: University of Western

Australia Press).Rutledge, M. (1979) ‘Barton, Sir Edmund (1849–1920)’, in Australian Dictionary of

Biography, vol. 7 (Melbourne: Melbourne University Press), pp. 194–200.Skidmore, M. J. (2004) After the White House: Former Presidents as Private Citizens

(New York: Palgrave Macmillan).Stewart, C. (2005) ‘Malcolm in the middle’, Weekend Australian Magazine,

27–28 August.Strangio, P. (2009) ‘Introduction: From confusion to stability’, in P. Strangio and

N. Dyrenfurth (eds) Confusion: The Making of the Australian Two-Party System(Melbourne: Melbourne University Press), pp. 1–19.

Switzer, T. (2008) ‘Former PM with manners of steel’, The Australian, 20 November.Theakston, K. (2006) ‘After Number Ten: What do former prime ministers do?’,

Political Quarterly, 77(4), 448–56.Updegrove, M. K. (2006) Second Acts: Presidential Lives and Legacies after the White

House (Guilford, Conn.: Lyons Press).van Onselen, P. (2009) ‘Howard’s “day job” a good fit’, Weekend Australian,

19–20 September.Walter, J. and Strangio, P. (2007) No, Prime Minister: Reclaiming Politics from Leaders

(Sydney: UNSW Press).Warhurst, J. (2008) ‘For Rudd, wayward Keating’s a player worth having on side’,

Canberra Times, 13 November.Watson, D. (2002) Recollections of a Bleeding Heart: A Portrait of Paul Keating PM (Sydney:

Knopf).Weller, P. (2007) ‘Prime ministers’, in B. Galligan and W. Roberts (eds) The Oxford

Companion to Australian Politics (Melbourne: Oxford University Press). Weller, P. and Fraser, S. (1987) ‘The younging of Australian politics or politics as first

career’, Politics, 22(2), 76–83.Whitlam, E. G. (1979) The Truth of the Matter (Melbourne: Penguin).Whitlam, E. G. (1985) The Whitlam Government 1972–1975 (Melbourne: Penguin).

Paul Strangio 229

13How Power Changes Hands:Concluding ReflectionsPaul ’t Hart and John Uhr

Introduction

Our aim with this volume has been to map out fresh initiatives in the study ofpolitical and administrative turnover which can test and refine existing inter-pretations of transitions and successions. The project is partly empirical andpartly theoretical. The empirical dimension is about extending the evidencebase and case material available to students of transitions and successions. Thetheoretical dimension is about assessing existing interpretations of politicaland administrative turnover against the findings emerging from this set offresh investigations of change in transitions and successions.

Our opening chapter framed the project by placing the chapters in thenarrower context of existing research on transitions and successions in gov-ernment. This concluding chapter draws on the empirical work in the preceding chapters to tease out the theoretical implications for the study of ‘power changing hands’ within political systems. Our hope is that wecan in turn stimulate further empirical and theoretical research building on our findings in ways that strengthen scholarly understanding of justhow such changes occur and how they play themselves out.

We carry out this concluding task in three steps. First, to clarify whereour project comes from, we revisit the two species of ‘power changing hands’captured in the project’s organizing categories of governing elite transitionsand leadership successions. Second, to build a bridge between where ourproject comes from to where we see it going, we pay particular attention tothe relevance of the qualifying term ‘government’ in our title, which definesthe broad field of social science interest that our project addresses. Third, inorder to specify the potential relevance of our findings to research on ‘gov-ernment’, we revisit the two sets of research perspectives relating to transi-tions and to successions which we sketched in our opening chapter. Herewe round out those earlier provisional sketches with two pictures of richerdetail (one on transitions and one on successions) drawn from the project’sevidence and findings.

230

Power changing hands

While the two shorthand terms of ‘transitions’ and ‘successions’ are useful,they are too general to capture the distinctive research preoccupations ofthe two sets of research investigations published in this collection underthose two broad headings. Not simply any transitions or any successionswere selected for inclusion in this project: the research cases were selectedon the basis of fitting a more targeted profile of governing elite transitionsand leadership successions. Let us therefore reset the focus on the largerstudy of power and of leadership, noting where transitions and successionsfit.

Governing elite transitions

The subject of ‘power’ is at or near the very centre of the social sciences.Power can be personal or institutional. Power is an important attribute of individuals, with many types of personal power conferring value andimportance on individuals. Power is also a vital social attribute: everysociety is organized around the management of power, with different typesof power defining different types of society, and with many different formsof power marking out different power-centres in each type of society.Power can be extracted, willingly or unwillingly, from its original pos-sessors or it can be imposed by the powerful on the relatively powerless.Power can also be dispersed or shared across many participating individualsor institutions, either to strengthen the overall capacity or to weaken thepotential for misuse by any one individual or institution. The patterns ofpower are important not only for so-called realist but also avowedly criticalstudies of society: power is the focal point of attempts to describe and to evaluate core social relations.

Concepts of power have themselves undergone transition and succession.The historical origins of modern society are associated with influentialtheoretical debates over power and how best to constitute the public man-agement of power. English political philosophers like Thomas Hobbes andJohn Locke are good examples of this concern over the challenging capa-cities of personal and social power which, if not properly managed, couldthwart the path of social and political development now associated withthe rise of modernity. The ambiguous nature of power as an importantpublic instrument for modern government emerges first in Hobbes, whoseLeviathan tries to demonstrate how power has to be centralized under theauthority of a sovereign armed with undivided coercive power to use forcarefully prescribed limited public purposes: most notably, the peace andsecurity of the public realm. The solution to the problem of endemic powerstruggles within society is to transfer power from society to government,suitably reconstructed as a representative institution to protect the interestsof those it represents. The more typically liberal blueprint of public power


emerges later in Locke who devises an influential account of liberal consti-tutionalism with ‘government’ broken up into three branches (executive,legislative, judicial) in order to promote civil liberties on the foundations of peace and security established by Hobbes. Public power is thus res-trained through constitutional separations of power exercised by rivalpower-holders.

Although political debate over power, elites and transitions of rulers goesback to classical times, modern social science interest is framed by theliberal-constitutional perspective derived from such influential thinkers asLocke. We think of transitions in terms that would have been familiar toLocke, who was preoccupied with questions of political legitimacy arisingfrom the rites of passage as governmental power passed through successivehands. Liberal-constitutional power is delegated power, with the peoplealone vested with sovereign power, some of which they delegate to theirchosen representatives. We can think of this as an original vertical tran-sition of power celebrated for example by Hobbes and Locke in the lan-guage and imagery of ‘the social contract’. The historical development ofliberal-constitutional regimes unfolds increasingly complex norms andpractices for what we might term horizontal transitions of power regulatingthe arrival and departure of each wave of legitimate holders of govern-mental power, with the rule of law stretched in different ways to cover the circumstances of the three increasingly independent branches of publicpower: the political executive, the legislature and the judiciary.

The ‘power transitions’ highlighted in this volume deal primarily with transitions in and out of the political executive: with the arrival anddeparture of newly elected governments. The main exception is the chapterby Uhr, Bach and Massiciotte which shows how transitions in executivebranches are often qualified by counter-transitions or non-transitions inlegislative branches. Liberal-democratic governmental powers are sharedpowers, as in this case of executive and legislative branches sharing publicpowers even to the point of affecting the transition plans of incominggovernments. But clarity of analysis usually means that researchers exam-ine power transitions in the one branch or department of government, typically the executive branch where relationships between political andadministrative executive officials provides fascinating case material for oursecond topic: leadership successions.

Leadership successions

Succession summons up pictures of ‘success’, as in the case of a successfulcampaign to displace departed, retiring, opposed or discredited power-holders. But the ‘success’ in question depends on the qualities of leadershipdisplayed first in the process of succession and later in the practices of thenewly-won leadership powers. Our interest in leadership successions is inpart an interest in the bridging process used by successors to move from

232 How Power Changes Hands: Concluding Reflections

‘outside’ to ‘inside’ (or ‘follower’ to ‘leader’, ‘downstairs’ to ‘upstairs’ andfrom ‘back office’ to ‘front office’). This dimension of our interest dealswith the displacement and replacement of predecessors. In some ways, it isa classic change-management story, where success is measured by the leastamount of agony and disruption caused by the process. But our interestalso addresses a second dimension of succession related to the leadershipcredentials of those undergoing succession (newcomers compared to thosebeing replaced) and in the wider leadership capacity that the successionmight be hoped to strengthen.

These two dimensions remind us that the study of leadership is notsimply the study of those with claims to lead. Their claims can be testedagainst standards of leadership, if we are prepared to construct flexiblestandards that fit the many different circumstances in which leaders func-tion. So too when examining leadership succession part of the task is tochart when, how and why replacements of individual power holders takeplace, and to reflect on what these patterns of succession convey. Whichleadership powers and capacities have been strengthened (or indeed weak-ened) through these succession processes? How have they altered the insti-tutional balance between the office of leader in question and those ofgovernment and party institutions around it? To what extent have theyaffected the legitimacy of these offices and the broader political system in which they are embedded? In short: a succession story is not solelyabout changing faces or changing places; it is also about evolving structuresand contexts for exercising public leadership.

The theme of leadership is related to that of power. Leaders exercisepower: leaders require power; and the public management of the powers ofgovernment requires leaders. Some public leaders display forms of leader-ship with largely informal powers, sometimes influencing government andsometimes bypassing the formal apparatus of state institutions. Examplesinclude those non-government leaders (these days including politicallyactive celebrities) exercising moral power which can influence the conductof willing followers and even, through its role in shaping community opinion,restrain the conduct of sceptical public officials in systems of supposedlyrepresentative government (Kane, 2000; ’t Hart and Tindall, 2009). Withthese cases of largely informal leadership, the transition process is likely tobe highly personal, testing claimants’ ability to compete for the mantle ofmoral authority of their predecessors. In this world of personalized leader-ship, qualities of personal character predominate in any succession process,where the power of ‘personality’ is a relevant test of succession. More tightlystructured are the transitions and succession processes in those front-stage,formal offices where leaders can wield very formal powers. As evident fromthe Cross and Blais chapter as well as the Laing and ’t Hart chapter, personalattributes and power struggles remain highly visible in party leadership tran-sitions. They are however more obscured and even downplayed in public


service successions, where they are routinely subordinated to ‘objectified’succession criteria such as policy skill and administrative competence.

Where is ‘government’ in this picture?

Our project has been explicitly about government. The examples of tran-sitions and successions chosen by our researchers have shared this focus ongovernment (stretched somewhat to also include successions within thepolitical parties that get to hold or participate in government from time to time). Government is one of the most prominent sites for exercising of public power. As mentioned above, the architecture of public power inmodern society deals with a fundamental dilemma facing modern govern-ment: how can the regulatory power of government be safely entrustedinto the hands of public representatives whose personal interests will notdivert the proper uses of those powers from the public benefits expected ofthem? Government must be empowered not for its own sake but for thesake of the public it represents.

This volume has addressed two important aspects of this liberal-constitutional architecture of public power. Our transitions researchershelp us probe into the value and trustworthiness of democratic proceduresfor managing the transition of modern government’s formal executivepowers from one political team to another. So too our succession researchershelp us assess different succession rules and practices for managing the safesuccession of holders of high political office inside or close to executivegovernment.

Government features most prominently in our set of transition stories.Simms carefully compares two national case studies of ‘caretaker con-ventions’ devised by Australian and New Zealand authorities to regulate the use of government powers during election periods. These are classicinstances of harm-minimization safeguards designed to restrict the abuse ofthe powers of government by political executives at times when their tenureis up for public reassessment. Errington and Walter compare leading oppos-ition and government figures in closely-related studies which demonstratethe function that checks and balances can have in managing expectations of incoming leaders about their powers and responsibilities. The study byUhr, Bach and Massiciotte examines the checks and balances arising fromseparated executive and legislative powers which potentially restrain the use of executive powers by incoming governments. Higley and Pakulski’sreflections on Pareto’s elite theory shows yet another balancing dynamic(though less of an explicit safeguard) at work in modern government in thealternation between two different types of governing elites identified byPareto in his language of lions and foxes.

Government is also present in the succession chapters, even if hiddenbeneath the surface of conventional accounts of governmental power. The


chapters by Cross and Blais, and by Laing and ’t Hart, on party leader suc-cession examine how political parties both inside and outside governmentcontinuously assess leadership performance and leadership potential withintheir own ranks. They do so through a remarkable array of potentially force-ful checks and balances that keep incumbent party leaders alert to the real-ities of internal competitive selection. Likewise, the Dowding and McLeaychapter compares two national experiences regarding the (im)prudenceexercised by prime ministers in managing both their governments and their senior party colleagues through a politics of (de)selection (i.e. cabinetaccession, promotion, demotion, and exit).

Government functions by virtue of its ability to exercise authority (poweraccepted as legitimate). Likewise, individuals arriving into high political oradministrative office within government will need to establish themselvesas wielders of that authority. Atkins’ chapter tackles the administrative sideof the equation by sketching the self-defeating nature of currently commonways in which incoming public managers assert themselves in their newroles, before outlining an alternative and more promising pathway. Subasicand Reynolds’ study of social-identify frameworks for analysing politicalsuccessions (as well as, potentially, transitions) shows how (government)leaders seek to establish and consolidate authority by narratives in whichthe boundaries and relations between ‘us’ and ‘them’ are selectively hard-ened or softened. They can try to portray themselves as ‘the one’ to save‘us’ from those who are not like us, or they can try to renegotiate who ‘we’ are in the first place, and play down rather than sharpen the salienceof some of the distinctions that exist between us and them. Depending on whether they read the mood of their intended followers correctly, suchpostures can make or break their own authority and that of the govern-ments they lead.

Finally, Strangio’s chapter opens up speculation about the end of theroad in succession, when the outgoing office-holder needs to let go of gov-ernment leadership. He notes that an increasing number of such formerheads of government will not just be willing but also young and fit enoughto somehow try to get back into the game of public leadership, either insideor outside official government or party roles. This growing reservoir of able-bodied, ambitious former leaders is, of course, a deeply ambiguousphenomenon: enhanced opportunities for key roles in managing the public’sbusiness open up; but also a potentially vociferous choir of critics withpotentially significant nuisance value for incumbent governments andtheir leaders.

Main findings

We can spell out some of the main findings from our chapters before we goon to tease out the larger implications for future research. Our first chapter


identified the main research questions tackled by the authors in the ‘tran-sitions’ and ‘successions’ sets of chapters. There our point was to note thevariety of research perspectives adopted by our research team, observingthat the set of five chapters on ‘transitions’ differ in research design fromthe set of six chapters on ‘successions’, and that within each set there aredifferent approaches and methodologies. We now highlight a set of findingsthat characterizes the collection as a whole to give readers a useful summarystatement of what the evidence from these research projects tells us about‘transitions’ and ‘successions’.

The transitions set of chapters provides surprising evidence about thedynamics of power changes when one political team replaces another.Students of democratic power transitions have a rich array of fresh evid-ence to draw on to broaden their hypotheses when trying to account forthe success and failure of different types of transitions. For starters, Higleyand Pakulski demonstrate that what we hold as ‘evidence’ about transitionsexpands and shrinks according to the theoretical perspectives we apply andthat Pareto’s analytical framework, when put to an empirical test, fore-grounds fascinating evidence of transition between two modal types ofruling elites. The evidence comes from their empirical comparison betweenthe USA and Australia, or between a large and diversified global power anda smaller middle power often working in international alliances with thelarger power. Does the evidence really confirm that the ruling politicalelites in both systems are transiting from leonine to vulpine governingmodes? Higley and Pakulski admit that it is too early to tell, but in manyways their main point is that none of this evidence of modal types in tran-sition would have come to light but for their unorthodox use of Pareto’ssocial theory of elite types as a guide to ‘what evidence to look for’.

Most of the other chapters in the transitions set are less explicitly theory-driven; they instead adduce evidence about power dynamics from empir-ical case studies of transitions. The researchers have in each case testedconventional wisdom by contrasting conventional practices of transitionwith exceptional or unconventional practices, hoping thereby to uncoverevidence to help explain why some transitions are more effective andothers less so. Thus, Walter compares models of leadership transition in hiscontrast between unconventional ‘managerial’ political transitions whichtend to survive through several elections and conventional transitions ledby political ‘messiahs’ which tend to fail at early electoral hurdles. Walter’sextensive comparison of Australian political transitions since the 1970sreveals the importance of evidence of the ‘personal style’ of incomingleaders, so that the exceptional ‘managerial’ types (e.g., Hawke) direct moreeffective transitions from opposition to government than do ‘messiah’types conventionally encouraged by structures of political opposition (e.g.,Whitlam). Errington in turn puts Walter’s focus on personal leadershipstyle to the test by producing evidence of the importance of structures of


parliamentary opposition to prepare exceptional opposition leaders for thetransition to government. Few opposition leaders make the transition fromleading the losers to leading the winners, and Errington’s group portrait oftwo UK and two Australian exceptional leaders is evidence of how priormastery of opposition resources is beneficial for later mastery of govern-ment resources, and of how this political skill first manifests itself in thetransition process.

Evidence of constitutional norms and practices is also important in ourresearchers’ contribution to the evidence about effective transitions. Thechapter by Uhr, Bach and Massicottte balances the chapter by Simms, withthe former dealing with evidence about ‘contested transitions’ under threeseparation of powers regimes and the latter dealing with evidence about‘caretaker conventions’ in Westminster-derived parliamentary systems.Both of these chapters uncover evidence about barriers and obstacles toeffective transitions which typically delay or frustrate new governments. Inthe former case, the evidence is that bicameralism complicates transitionsby providing opposition parties a safe refuge where they can share controlover legislative power while conceding executive power to the new gov-erning party. In the latter case, the evidence is that constitutional normsregulating ‘caretaker conventions’ prove increasingly contentious for gov-ernments when approaching elections, as opposition challengers try to ‘levelthe playing field’ of party politics by enforcing rules that rob governingparties of some of the core advantages of incumbency. The evidence fromboth chapters shows how hard (e.g., written and entrenched) and soft (e.g.,flexible guidelines) constitutional structures affect the way we define tran-sitions and the ways that political leaders compete for more than simplyexecutive power: for legislative power where that is separately available inone or more branches of the legislature, and for caretaker powers wherethere are constitutional conventions curtailing the available power at thedisposal of governing parties.

The set of chapters on succession is intriguing in that they offer differentbut complementary vantage points and loci of analysis for understanding(and evaluating) succession practices. The Atkins and the Strangio chapterhelps us understand succession dynamics solely through the prism of theindividual leader, whether it is a new (Atkins) or an outgoing (Strangio) one.In their analysis, it is the leader’s personal capabilities, drive and style thatshape their fate in the post-succession context. Atkins focuses on the newleader’s ability to remain reflective, open-minded and empathic towardsstaff and colleagues, instead of shooting straight into decisive, ‘I-know-best’‘action’ mode. Strangio focuses on the resilience former leaders need tocarve out a meaningful post-executive existence. Their personal health,ambition and flexibility play an important role in this, but he also pointsout that contextual (the former leader’s standing in their party and in gov-ernment circles more widely) and institutional (pension and allowances


factors feed into their post-prime ministerial career choices and desires toremain in (or to disappear from) the public sphere.

In the Dowding and McLeay chapter, the focus is also squarely on oneindividual, though not the leader whose position is at stake in successionepisodes. Their focus is on the role of prime ministers as decision-makersabout the careers of the members of the cabinets they lead. But their studyof British and New Zealand practices demonstrates clearly that prime ministers are hardly free agents in this regard. Their choices in managingcabinet composition are shaped to a considerable extent by forces of cir-cumstance: the factional power or other political clout of ministers (e.g.‘Big Beasts’), media criticism and parliamentary ‘flack’ that ministers – andtherefore the government as a whole – are enduring, and of course the exigencies of the electoral cycle. The principal-agent perspective thatDowding and McLeay present in capturing some of these dynamics is a useful reminder that all senior public office-holders that we commonlyrefer to as ‘leaders’ are in fact always embedded in webs of hierarchy, repre-sentation or reciprocity. They can be held to account by more senior leadersor bodies – including by means of involuntary rotation or dismissal – whentheir actions become a focus of political debate.

The Laing and ’t Hart chapter focuses on the fact that the fates ofincumbent, aspiring and outgoing leaders are closely intertwined in a tact-ical game of coexistence, cooperation and at times open conflict withintheir parties and governments. Their dataset study makes it abundantlyclear that any attempt to build a presently non-existing theory of leader-ship succession should take into account the dynamic interplay betweenthese individuals and the party factions and power brokers aligned tothem. In the Reynolds and Subasic chapter, the focus is also on relationsrather than individuals, but the relation they argue that is most pivotal is that between leaders (whether aspiring or incumbent) and the groupsthey purport to lead. The only other chapter in the succession set besidesDowding and McLeay’s whose analysis is grounded in an explicit theo-retical framework (in this case social identity theory), it introduces a com-prehensive framework for understanding the success and failure of leader’sauthority claims. Turning many conventional notions about leadership on their heads, the social identity approach claims that people can onlybecome leaders when they successfully embody and appeal to the momen-tarily salient dimensions of ‘we-ness’ in the groups from which they emerge.However, Reynolds and Subasic also show that individual leaders have con-siderable scope for ‘identity entrepreneurship’ – including demonizing ‘out-groups’ and otherwise fostering exclusivist notions of ‘we-ness’ – in raisingor dampening particular types of identity considerations. Though tentativeand exploratory, their chapter puts the psychological contract betweengroups and their leaders squarely on the agenda of succession (and indeedtransition) research.


Finally, the Cross and Blais chapter goes beyond psychological contractsto look squarely at the role of institutional rules of the game governinghow groups (in their case political parties) select and dismiss their leaders.These rules formalize some of the principal-agent dynamics also discussedin the Dowding and McLeay chapter: they allow (elites within) parties tocontrol those who are nominally put ‘in charge.’ These rules continue toevolve within political parties, with a firm though not universal movetowards ‘democratization’ being visible in recent years. This chapter usesevidence from parties in five Westminster countries to demonstrate howsuch rules make for remarkably different degrees of ‘job security’ amongparty leaders. Selecting leaders in is one thing; selecting them ‘out’ can be quite another. The Cross and Blais chapter shows us that we should look at both sides of the succession coin, with leader removal perhaps the most sensitive issue (witness the ruthless way in which the parliamentaryAustralian Labor Party got rid of its hitherto extremely successful but inter-nally disliked leader and prime minister Kevin Rudd in June 2010). The keyvariables in this regard are: whether a formal leader removal mechanismexists at all; whether the extra parliamentary party must in some way act toremove the leader; and when the parliamentary party has this authority,whether it is in any significant way encumbered in its use.

Taken together, the studies reported in this volume map out a broad terrainempirically as well as conceptually. In the final section of this chapter, wereflect on the roadmap for future research that may be drawn on the basisof the work presented here.

Transition and succession research avenues

By way of conclusion, we present some lines of inquiry beyond the currentproject. Revisiting the analytical perspectives identified in the introductorychapter in light of the findings of the chapters that followed providesimportant clues to transition and succession research priorities.

Constitutional perspectives on transition

Structures of government regulate many aspects of governmental transition.Regular popular elections provide most political communities with oppor-tunities to have their say about the elected representatives competing for government office. But most constitutional arrangements leave many of the most important practical decisions over who wins and who loses in government transitions processes to the transitory team of politicalinsiders, particularly party-political power-holders. Few liberal-democraticconstitutions attempt to do very much to regulate the internal operationsof political parties which are, in many ways, the really important powercentres of representative government. The participation of voters is oftensubject to more extensive government regulation than the conduct of elected


and administrative party officials who can use their considerable insiderpower to change the electoral team at the top of party tickets. Even heads of governments are under the watchful eye of backroom party operatives whosee their role as ‘protecting the brand’ of the party against any diminishingreputation that a leadership team might suffer.

Simms notes that prime ministers have become increasingly strongdrivers of transition processes. Another crucial area for further explorationtherefore is the extent to which the political calculus of prime ministerscan come to dominate and indeed compromise the integrity of the publicservice’s obligation to safeguard the continuity of administration in a neu-tral and restrained manner during the caretaker period (whilst at the sametime feverishly preparing parallel sets of highly politically sensitive briefsfor the post-transition period).

Leadership perspectives on transition

How do aspiring heads of government prepare themselves, their parties andtheir publics for transition? To what extent are their pre- and post-transitionpostures shaped by their broader political personality and their socializationinto leadership roles? And, perhaps most fundamental of all, to what extentare the leaders of election-winning parties actually able to put their personalstamp on the substance, composition and style of the new governmentsthey front up?

Errington’s and Walter’s chapters suggest that such questions are worthexamining in greater detail, both within and across polities. On their accounts,the answers to these questions appear to vary considerably from leader toleader and transition to transition. Whether this variation is random or patterned across parties, time, and political systems is one puzzle that sys-tematic comparative research should address. Also, we need more systematicexamination of Errington’s assertion that opposition experiences are impor-tant predictors of prime ministers’ early modes of operation. The same goesfor Walter’s claim that there probably is some correlation between primeministerial transition styles and the success if not the longevity of the gov-ernments they lead. Such research should reach well beyond their focus on Westminster, single-party or fixed-coalition governments, and open upto the possibility that the imperatives and options for prime ministers ofmulti-party coalition governments common in polities with proportionalrepresentation electoral arrangements.

Executive government perspectives on transition

A broader executive government perspective on transition requires goingbeyond the contributions to this volume and examines the dynamics of political-administrative relations during the post-transition ‘settling in’ pro-cess of a new government. Even in the Westminster system with its allegednorms of bureaucratic impartiality and professionalism (in self-conscious


contradistinction to the ‘government of strangers’ produced by the Americanspoils system, cf. Heclo, 1976), new political masters can choose to change the rules of the game governing the relationships between themselves and the agencies and administrators in their portfolio.

For one, they can choose to retain or dismiss top executives. In Australia,for example, in 1996 the incoming liberal Howard government sacked six department heads within a week of taking office. In late 2007 the arrivalof its successor, the Labor Rudd government, was therefore anticipatedwith trepidation by the senior cadres of the public service. To great relief,Rudd chose to publicly emphasize their adherence to the ‘Westminster norm’of placing trust in the competence as well as the responsiveness of the permanent bureaucracy they had inherited.

Both opposing styles of dealing with this inheritance constitute signalsthat new governments send to the bureaucracy. And as always in the gameof communication, the medium is the message: does the new governmentspeak softly, or does it wield its big stick when it comes to deciding the fateof the leaders of the ‘permanent’ bureaucracy? With much of the attentionfocused on the dramas of personalities and careers, it is easy to forget thatnew governments possess various other means of reshaping the politics-administration nexus. They can change key procedural rules of the govern-ment game: cabinet submission routines that bureaucrats need to observe.They can install, change or scrap the contents or indeed the existence ofpublic service performance agreements. They can limit or maximize theaccess that public servants enjoy to them. They can appoint political streetfighters or collaborative policy wonks as their chiefs of staff. What theycannot do, is to completely ignore their strong dependence on the perfor-mance of the bureaucracy. At the end of the day, governments get judgedto a significant extent by the results they achieve; and the bulk of the deliverythat makes or breaks those results will always be done or managed by theirbureaucracies.

So both parties have a lot riding on making a productive start to the newworking relationship. There is however no systematic comparative study ofthe principles and practices by which new political rulers ‘join up’ withexisting bureaucracies, and their effects on the productivity of their sub-sequent collaboration. Such research is much needed, not in the least from apractical perspective. When speaking off the record, many bureaucrats con-fess that their common mode of receiving new governments is somewhatstale and definitely suboptimal. They note that the traditional emphasis on providing piles of briefings, thorough options papers and business-likeeagerness to ‘get on with the job’ under the new management, studiouslyavoids the pivotal dimension of interpersonal trust-building across whatmay well seem to both parties to be an ocean of differences. Westminsternorms notwithstanding, for potentially cagey, inexperienced new govern-ments, the sheer number of officials they are now responsible for and the


knowledge that all of them have worked for the previous government maybe a major impediment to according the trust bureaucrats know they cannot do without.

Something similar can be expected under the radar that protects govern-ment bureaucracy from intrusive public gaze. What needs to be known nextis the extent to which the new configuration of outside forces (politicians,their mates, peak bodies and other organized interests) that is formed on thewings of transition, modify not just the bureaucracy ‘authorizing environ-ment’ but also affect its very internal composition and ‘operational capacity’(Moore, 1995). The standard picture is one of merit-protected processescomplying with norms of strict impartiality, counterbalanced by an expect-ation of responsiveness to the (new) ‘government of the day’. Impression-istic evidence of recent transitions suggests limits to this picture, as variousforms of favouritism and partisanship have led some to fear that more andmore, political forces conducing towards bureaucratic ‘hyper-responsiveness’are crowding out Westminster traditions of impartiality and professionalism(Mulgan, 2007). The question is whether such impressions are borne out by larger-N comparative study, and if so, what can be done to ameliorate thissyndrome.

Power perspectives on succession

Studies of succession often differ in one important respect from studies oftransitions. Both deal of course with choices about the future. Transitionsmostly have regular and predictable timetables, such as electoral dates. Suc-cessions are less predictable and more insidious: they can strike almost atany time. Many deposed leaders ended up as such by complete surprise to themselves and their supporters.

Transitions are competitions for authority; successions are as well, butthey are more fundamentally competitions for power. Transition studies ofgovernment can show the phases in the move from opposition into publicauthority without having to explain the origins of the power that enabledthe move into government. Succession studies in contrast are more mind-ful of the rough and tumble of Roman or Renaissance style court politics.They need to come to terms with the over-the-shoulder anxiety by power-holders about ‘who’s moving up’, the speculations and the plots, and thebrooding presence of ambitious leadership aspirants biding their time untilthe time to strike seems right. They deal with the claims and machinationsof the people who carry the case for the party or institution obtainingpower. At the very root of that story is some explanation of how those indi-viduals (plan to) use the resources at their disposal. We will know moreabout successions in political and governmental organizations when wecan better explain why some individuals or groups are more successful inmobilizing and managing the power that makes their succession planssucceed.


Laing and ’t Hart note that political leaders engage in a delicate balanc-ing act in order to survive and thrive. The classic twin imperatives of publicpopularity and high standing in the party do not necessarily align, yetmaintaining both is needed in order to effectively lead. They observe thatwhilst conventional party wisdom might identify the perfect successor as one who stresses continuity and loyalty with the former leader, hasstrong and abiding links within the party, and succeeds his predecessorunchallenged and universally accepted, the findings of their comprehensivedataset study suggests that when parties get restless in the face of politicaladversity, a maverick populist who emphasizes a clean break with party policyand takes down his predecessor in a party-room showdown might have just as much (indeed, if not more) chance of successfully leading the party intothe future. With their findings – and with Subasic and Reynolds’ intriguingmodel of the role of identity entrepreneurship in power consolidation in mind – studying more carefully how government leaders win, preserve, and lose, their positions and their political capital is therefore an importantresearch domain.

Accountability perspectives on succession

Democratic leaders are accountable for their exercise of public power. Studiesof succession show how this accountability operates: how it is managedand how it can be subverted, by leaders and by the constituencies andforums that select and empower them but may also curtail them or evenstrip them of their offices. Cross and Blais’ chapter is a prime example, inits examination of the extent to which the prevailing rules for selecting anddeselecting party leaders enable MP’s, party elites or ordinary members to hold those leaders to account. They blaze a trail worth following, as the connections between various (de)selection mechanisms and key party/government leader attributes such as their authority, responsiveness andlongevity are as yet ill understood.

Above and beyond their focus on formal party mechanisms for leader(de)selection, we need also to track the informal, and partly invisibleaccountability dynamics governing leadership succession. We need studiesof the hidden claims of obligation leaders owe to influential interests in the backroom of any organization (e.g. the ‘big beasts’ noted in Dowdingand McLeay’s chapter, but also non-cabinet party power brokers, campaignfinanciers, bosses of organized interests or, these days, celebrity endorsers).Such studies should go ‘inside’ the politics of party leadership in order to help us grasp and asses the significance of these vital inside, informaland invisible dimensions of accountability (as compared to that of outside,formal and transparent ones) in shaping the coming and going of politicalleaders and the careers of ministers. As noted in the introduction, there are some gripping single case accounts of this variety written by politicaljournalists and historians (Reid, 1969; Koerfer, 1987[1998]), but there is as


yet no systematic political anthropology of party leadership or prime ministerial leadership.

Towards evaluation?

Finally, we acknowledge the need for more explicit and systematic assess-ment for both succession and transition regimes. That original manual insuccession management, Machiavelli’s The Prince, blended short- and long-term outcomes in its measurement of success in acquiring power: short-term goals like replacing existing rulers and long-term goals such as theglory that can come from using newly-won powers for great accomplish-ments. Machiavelli’s The Prince is also revealing in another way: this workintroduced the analytical study of power politics into the front rank ofpolitical studies and thereby overhauled traditional evaluation frameworksused in the study of government and politics.

The Prince dramatized politics with its case studies of political theatre andthe ways of stage-managing political succession. Machiavelli’s work is repletewith valuable case studies of succession that remain dramatically rewardingfor those searching for ways of evaluating succession as political perfor-mance. But as we reported in our introductory chapter, democratic legit-imacy might demand stricter standards of transparency and inclusivenesswhich, we can only hope, produce modes of succession that measure up assustainable democratic expressions of legitimate leadership selection. Variouschapters in this collection contain implicit judgements about the effectiveness,legitimacy or desirability of various modes or styles of transition/succession(e.g. Walter; Simms, Cross and Blais, and particularly Atkins). Laing and ‘t Harteven follow the example set in the opening chapter and attempt to offer criteria sets to be used in evaluating successions, suggesting they should beassessed in terms of their effects on policy, party popularity and regime legit-imacy. That said, the contributions in this volume already stop short of deliv-ering on the empirical question concerning what consequences differentmodes of transition and succession may have. And they are far removed fromengaging systematically with the thorny normative question which mode oftransition or succession is to be preferred over others in which institutionsand/or sets of circumstances. Until we take up this challenge, our ability to provide practical advice about transition and succession managementremains constrained. Filling this void is perhaps the most important but alsothe most challenging of the avenues for further research that emerges fromthis volume.

Bibliography

’t Hart, P. and Tindall, K. (2009) ‘Leadership by the famous: Celebrity as politicalcapital’, in J. Kane, H. Patapan and P. ’t Hart (eds) Dispersed Democratic Leadership:Origins, Dynamics and Implications (Oxford: Oxford University Press), pp. 255–78.


Heclo, H. (1976) A Government of Strangers (Washington, D.C.: Brookings).Kane, J. (2001) The Politics of Moral Capital (Cambridge: Cambridge University Press).Koerfer, D. (1987) [1998] Der Kampf ums Kanzleramt (Munich: Propylaen).Moore, M. (1995) Creating Public Value: Strategic Management in Government (Cambridge:

Harvard University Press).Mulgan, R. (2007) ‘Truth in government and the politicization of public service advice’,

Public Administration, 85(3), 569–86.Reid, A. (1969) The Power Struggle (Sydney: Shakespeare Head Press).


Abbott, Tony, 32, 186, 187accountability perspectives, on

succession, 13, 15–16, 243–4see also succession

ACT, New Zealand, 140, 144, 145, 152,172n5

adverse selection, 159–60New Zealand experience, 168–70UK experience, 162–4

agency rent, 158–9, 170New Zealand experience, 164–8UK experience, 160–2

agenda-setting, 58, 59, 66agent-shirking see shirkingaspirants, 116–21Australia, 6, 8, 95, 101, 134, 136, 143,

152, 174, 175decision-making units in Cabinet

system, 40executive government perspectives

on transition, 241framework of constitutional

government, 86–91executive power, 87judicial power, 87legislative power, 87–8

leonine ascendancy in, 31–3opposition leadership and, 58prime ministerial afterlives in see

prime ministerial afterlives, inAustralia, 208–26

rules for restraining incumbency,101–5

senatorial transition in, 88–91Australian Labor Party (ALP), 87, 102Australian Public Service

new leadership role, demands of, 192

authentic leadership, 196–9authentic behaviour, 197–8balanced processing, 198relational transparency, 197self-awareness, 198see also leadership

authority, 12, 15, 177, 178democratic, 2power transition in, 9

Barton, Edmund, 210income, 215influence of, 213

Beazley, Kim, 44, 102behavioural integrity, 197–8bias

availability, 195confirmation, 195

bicameralism, 11, 73–92big beasts, 159–60, 162, 163, 164,

169–70Blair, Tony, 136, 161–2, 174, 184

learning to lead in opposition, 71nationalism, 67, 68New Labour project, 62opposition leadership and, 55, 57, 59political learning in opposition

influence, 64, 65–6Bloc Québécois, 85, 140, 144Bolger, Jim, 135, 150, 158, 165, 166, 170Brash, Don, 150, 151Bringing Them Home report (1997), 68Brown, Gordon, 45, 70–1, 149, 162, 163,

169, 174policy fiefdoms, 64

Bruce, Stanley Melbourne, 211coming back to Australia, 214as High Commissioner, 214as inaugural chancellor of the

Australian National University,214

income, 215influence of, 213international diplomacy, 214representative at League of Nations,

214Bruton, John, 136, 151, 154n5bureaucracy, 6, 12, 15

authorizing environment, 241–2hyper-responsiveness, 242

246

Index

Bush, George W., 45, 175, 187administration, 29–30

butchers, 159, 160, 166, 171

Cabinet Government, 96, 97, 99Cabinet Manual, 99, 100cabinet ministers, 12, 14, 15, 16Canada, 80–6, 134, 136, 137, 152

Charlottetown Accord, 82Constitution Act of 1867, 84Meech Lake Accord, 81new government, options for, 83–6Penitentiary Act, 82senatorial transitions in, 82senators, options for, 82–3

CARE Australia, 224Malcolm Fraser and, 225

caretaker conventions, 11, 94, 237caretaker periods, and management of

transitions, 94–5checks and balances, 4, 234, 235chief executives, 13, 14, 15Chrétien, Jean, 82, 83, 135, 149–50Clark, Helen, 99, 101, 103, 106, 164,

165, 167, 168, 169, 170Clinton, Bill, 77

administration, 29, 34Coalition Government, 3, 95collaborative action planning, 202–3collaborative problem solving, 197collaborative sense-making, 202Collective Responsibility, 98–9Committee of Officials, 99Commonwealth Electoral Act (CEA),

102, 103conflicted successions, 126–7

see also successionsconsensus seekers, 159, 161, 167,

171Conservative party, UK, 135, 136,

137, 140, 144, 149, 151, 152consociational democracies, 11constitutional perspectives, on

transition, 10, 239–40consultative leadership, 197

see also leadershipcontested transitions, 73–92, 237

Australia, 86–91Canada, 80–6dispersed power and, 91–2United States of America, 76–9

see also transitioncontextual factors, 7Conventions, 94Cook, Joseph, 211

coming back to Australia, 215influence of, 213–14

Costello, Peter, 70, 174Costigan Royal Commission, 104cronyism, 161–2Curtin, John, 87, 210

database, 112dataset, 9, 17, 112, 114, 115, 118,

120, 122, 124, 127, 128Deakin, Alfred, 211, 220

influence of, 213poor health, 212

Defence of the Realm Acts, 98demagogic plutocracy, 26

see also plutocracydemocracy, 4, 6, 12, 17, 94, 112

consociational democracies, 11degrees of popular control, 4–5democratic transitions, 3–4,

9–10parliamentary democracies, 2, 8,

13and peaceful transition, 1–2working democracy, 1

democratic consolidation, 4democratic leadership, 111democratic transitions, 3–4, 9–10

see also transitiondialogue, 201–2Diefenbaker, John, 82, 83, 124, 136,

137, 148dispersed power, 91–2divided government, United States

of America, 77–8division, politics of, 181, 182, 184dominators, 157, 159, 161, 162, 165,

167, 169, 171Douglas, Roger, 164, 165, 168, 170,

172n5

elections, 1, 2, 3, 4, 10, 11, 17electoral cycles, 8electoral impact, 114electoral process, 1–2electoral systems, 2, 95electorate, 1, 2, 9

Index 247

elite(s)circulation, 25degeneration, 25–6and electoral competitions, 29–30leonine, 27–33

ascendancy in Australia, 31–3ascendancy in United States, 28–31

propensities, 27ruling, 23transition, 23–5, 231–2Pareto’s theory of, 24–7ruling elites, 26–7types, social theory of, 236vulpine, 33–5see also transition

emotions, 193empire, 214–15EntryPlan, 199–203

collaborative action planning, 202–3collaborative sense-making, 202designing, 200–1gathering data through dialogue,

201–2testing, 201

Europhiles, 164Eurosceptics, 164evaluation, 16exclusion, 176, 178, 179, 180, 181executive government perspectives, on

transition, 2, 3, 6, 12, 15, 240–2see also government

executive power, 4, 10, 11

factional rivalries, 2Fadden, Arthur, 210, 217federalism, 91Federated Ships Painters’ and Dockers’

Union, 104Fianna Fáil, 135, 140, 145Fightback!, 48, 62, 70fight-flight response, 193filibuster, 80financial crisis, 45Fine Gael, Ireland, 136, 137, 140, 141,

144, 151, 152, 154n4firing of ministerial colleagues, by Prime

Ministersagency theory perspective, 158–60moving ministers, UK and New

Zealand experiences, 157–8in New Zealand, 164–70

Prime ministers and ministerialturnover, in Westminster, 170–1

in United Kingdom, 160–4Fisher, Andrew, 220

coming back to Australia, 214–15income, 215influence of, 213poor health, 212

forced exits, 160Fraser, Malcolm, 42, 87, 105, 210, 217

CARE Australia and, 224as chair of the United Nations

Secretary’s Expert Group onAfrican Commodity Issues, 225

Commonwealth Group of EminentPersons against Apartheid inSouth Africa and, 225

defeat of, 48as Liberal leader, 217moving on and filling the void, 223,

224opposition leadership and, 57raging against successors, 221–2Robert Menzies and, 217self-justification, 222transition to government, managing,

46–8Future Directions, 50

Goods and Services Tax (GST), 62, 82Gorton, John, 135, 210

raging against successors, 221removal as Liberal leader, 217self-justification, 222

government, 234–5ability to exercise authority, 235branches of, 232executive, perspectives on transition,

240–2regulatory power of, 234

during election periods, 234stability, 16successions within political parties,

234–5Governor-General, 102, 104Green Party, Ireland, 144, 145Green party, New Zealand, 134, 140,

144, 145

Hanson, Pauline, 49, 68Haughey, Charles, 135, 145, 146

248 Index

Hawke, Bob, 28, 38, 42, 69, 101, 103,106, 210

in election 1983, 42Hawke Research Institute and, 224laissez faire approach, 43policy development, 42self-justification, 222transition to government, managing,

42–3, 51–2Heath, Edward, 136, 154n3, 161, 162heroic leadership, 192

see also leadershipHeseltine, Michael, 161, 162Hewson, John, 31

critique on Howard, 49Fightback!, 48, 62, 70opposition leadership and, 59

high-conflict succession, 126–7Hobbes, Thomas, 231–2Holland, Sidney, 158, 164, 167Holt, Harold, 210, 216Holyoake, Keith, 167–8, 169Howard, John, 38, 89–90, 95, 101, 103,

104, 174, 175, 180, 181, 184, 187,210, 219, 241

Bringing Them Home report 1997 and, 68

criticized by Hewson, 49economic transformation and, 61–2Future Directions, 50government, 31–2learning to lead in opposition, 71Mark Latham and, 63moving on and filling the void, 223nationalism, 68opposition leadership and, 55, 57,

59, 60political learning in opposition

influence, 64–5presidential campaign, 48raging against successors, 221self-justification, 222transition to government, managing,

48–9, 51Hughes, William Morris, 210, 212,

217income, 215

identity entrepreneurs, 178identity entrepreneurship, 12, 238, 243inclusion, 179

income, 215–16incumbent, 3, 5, 7, 8–9, 11

leaders, 121–4Indigenous – Non-Indigenous

relations, 185influence see persuasioningroup, 176, 177, 178, 181, 184

solidarity, 179, 180victimization, 180

intergroup, 176, 177, 181, 182, 183conflict, 179

interparty competition, 181, 183intragroup, 176, 177, 181, 182, 183

solidarity, 186intraparty competition, 181, 183Ireland, 134, 136, 152

Keating, Paul, 37–8, 42, 44, 62, 63, 70–1, 102, 103, 104, 127, 187, 210

government, 31raging against successors, 221, 222self-justification, 222as visiting professor of public policy

at the University of New SouthWales, 224

Kennedy, John F., 39, 179Kerry, John, 30, 87

Labor and quality of government, 41Labor governments, preparation and

transition, 38–46Labor party, Australia, 135, 138, 151Labour party, Ireland, 138, 144, 145Labour party, New Zealand, 95, 97,

103, 135, 136, 138, 145, 151Labour party, UK, 136, 137, 141, 145,

149, 151laissez faire approach, 43Lange, David, 135, 158, 165, 168, 170Latham, Mark, 220

John Howard and, 63leader

cognitionsattention, 194private versus collaborative

information processing, 194–6longevity, 114reputation, 196, 213

Leader Democracy, 23leader-follower relations, 176

Index 249

leadershipaspirant, 116–21authentic, building trust through,

196–9challenge, 118, 122, 123, 127change

power consolidation and socialidentity and, 178–81

as social identity, 175–8consolidation, 12consultative, 197heroic, 192longevity, 112, 113opposition see opposition

leadershippersonal, 236–7perspectives on transition, 240rival, 8selection, 15, 114, 121, 122, 129selectorate, 151–2strong, 56, 64struggles, 2, 7style, 11, 14, 56–7successions, 5–7, 232–4

non-existing theory of, 238see also successions, 232–4

transactional, 57transformational, 57see also new leadership role

legitimacy, 177, 178, 181leonine elites, 27–33

ascendancy in Australia, 31–3ascendancy in United States, 28–31see also elite(s)

Liberal Democrats, UK, 141, 144, 151,152

Liberal governmentspreparation and transition, 46–50

Liberal party, Australia, 135, 138, 150,151, 186

Liberal party, Canada, 135–6, 137, 140,141, 144, 151, 152

Liberal People’s Party (LPP), 123Lijphart, Arend, 112Locke, John, 231, 232

Machiavelli, Niccolò, 244Major, John, 117, 161, 164Maltese Labour Party, 113mandate, 69–70Mandela, Nelson, 130

Maori party, New Zealand, 134, 138,144, 170

Marshall, John, 158, 169, 170McCain, John, 23, 31McMahon, William, 210, 218

moving on and filling the void, 223raging against successors, 221

media, 96, 209, 219, 220management, 65

Meech Lake Accord, 81memoirs, 217–18Menzies, Sir Robert, 209, 210

Malcolm Fraser and, 217memoirs, 217–18transition, 216–18

mergers, 2military plutocracy, 26

see also plutocracyministerial resignations, 158, 160, 161,

162–3, 164, 165, 166, 168, 169, 170

ministerial successions, 5, 13see also successions

ministrables, 158, 159, 164, 167, 168,169

minority government, 3Mixed Member Proportional (MMP)

electoral system, 95, 96, 98, 99Model 1 behaviour, 194, 195Model 2 behaviour, 194–5Moore, Michael, 151, 158Muldoon, Sir Robert, 99, 164, 165, 168,

169, 170Mulroney, Brian, 82, 84, 127

National Government, 100nationalism, 67–9National party, Australia, 138National party, New Zealand, 95, 99,

135, 136, 140, 150, 165, 166, 167,170, 172nn5, 6

Nelson, Brendan, 32, 150, 174, 175, 186Neustadt, Richard, 39New Democratic party, 144New Democrats, Canada, 140new leaders, 8, 12, 16, 17, 116–21new leadership role, in public service

approach to collaborative enquiry forsee EntryPlan

demands of, 192–4objections and risks, 203–5

250 Index

private processing of information,194–6

public sector managers, 191–2short-term demands versus long-term

interests, 195–6trust through authentic leadership,

building, 196–9see also leadership

New South Wales Traffic AdvisoryCommittee, 216

New Zealand, 6, 8, 11, 16, 95, 134, 136,143, 152, 153, 158, 164–70

rules for effective electoral change,98–101

New Zealand First, 140, 166non-democratic systems, 9–10non-events, 9norms, 177, 178, 180, 188

Obama, Barack, 23, 31, 175, 179, 183,184, 185

administration, 31, 33–4office-holders, 4, 5, 12, 13, 14, 209,

219, 235open communication, 199operators, 159, 160, 161, 166, 167,

171opposition, 2, 3, 11opposition leaders, 11opposition leadership, 55–72

as a formative experience, 58–9experience, governing without,

69–71learning, 71–2learning to manage government,

63–7learning to speak for the nation,

67–9tools of, 59–61winning formula, 61–3see also leadership

organizational learning, 194–5outgroups, 12, 176, 177, 178, 184, 238

exclusion, 180, 181marginalization, 180

Page, Earle, 210, 217Palmer, Geoffrey, 135, 156n6, 158, 166,

170Pareto, Vilfredo, 23

theory of elite transitions, 24–7

parliament, 2–3, 11, 16, 94, 210, 215parliamentary democracy, 2, 8, 13

parliamentary democracies, 2, 8, 13parliamentary majority, 3parliamentary sovereignty, 97parliamentary supremacy, 97parties, 2–3, 4, 5, 6, 9, 10, 11, 13, 16,

17party leaders, 5, 12, 13, 14, 15, 17

party’s first post-succession election,113

party leader, 111, 112, 113, 117, 118,120, 125

job security among, 239removal of, 239

party organizations, 16party room, 112, 123, 125, 128, 131n4party rules, 129patronage, 13Peacock, Andrew, 61, 104, 151Penitentiary Act, 82pensions, 215, 223, 225personality, 233perspective-taking, 195persuasion, 178, 213–14Peters, Winston, 165, 166, 167, 170plutocracy, 26–7

demagogic, 26military, 26

policy shifters see shifterspolitical learning

to manage government, 63–7to speak for the nation, 67–9

political parties, 2, 4, 5, 6, 10, 13, 17political structures, 6political transitions, 1

see also transitionpostelection period, dealing with

uncertainty in, 95post-leadership, 209post-succession election, 114power, 14, 15, 175, 177, 188

brokers, 116changing hands, 231–4consolidation, 178–81, 183, 184, 186,

187liberal-constitutional, 232liberal-democratic, 232perspectives on succession, 242–3separation of, 237

Prebble, Richard, 165, 168, 170, 172n5

Index 251

predecessor and successor, conflictbetween, 120–1

presidentialization, 15presidential systems, 3, 13prime ministerial afterlives, in Australia,

208–26coming back to Australia, 214–15income, 215–16influence, 213–14modern era, 218–20moving on and filling the void, 223–5pre-modern era, 211–12raging against successors, 220–2self-justification, 219, 222

prime ministerial successions, 13prime ministerial transition

management, 37–53Labor’s experiences, 38–46Liberal approaches, 46–50styles of, 37–8, 50–3

prime ministers, 2, 11, 12, 13as decision-makers, 238, 240leadership style, 14, 17

in opposition see oppositionleadership

see also butchers; consensus seekers;dominators; operators

modes of operation, 240power of, 15as primus inter pares, 209

principal agent see agency theoryprivate processing of information,

194–6Productivity Commission, 49Progressive Conservatives, Canada, 136,

137, 140, 141, 142, 148Progressive Democrats, Ireland, 144,

152, 154n4proportional representation, 2, 6, 88public administration, 4public sector managers

new leadership role, 191–2public service, 2, 6, 8, 12, 14, 15, 94public service successions, 6

see also successions

Quigley, Derek, 164–5, 166, 172n5

Reagan, Ronald, 78–9administration, 28–9

recall Parliament, 101

re-categorization, 176Reid, George, 211, 217

coming back to Australia, 215income, 215influence of, 213–14

relational transparency, 197relative prototypicality, 124removal, 15, 16, 239reshuffles, 2, 5, 168–9responsiveness, 5, 12, 13, 15revolutionaries, 120Rowling, Bill, 145, 158Rudd, Kevin, 23, 89, 175, 185, 241

global financial crisis 2008 and, 45government, 32–3, 34–5nationalism, 68–9opposition leadership and, 55, 57political learning in opposition

influence, 66–7transition to government, managing,

44–6, 52winning formula, 63

ruling elites, 23, 24circulation, 25degeneration, 25–6psychological propensities for, 24–5transition, 26–7see also elite(s)

Scottish National Party, 154n1Scullin, James, 210, 211, 220

income, 215influence of, 213poor health, 212

selectorate, 9, 111self, 176self-awareness, 198self-categorization theory, 176self-justification, 219, 222self-management, 198self-regulatory capability, 198Senates, 73–4

contested transitions inAustralia, 86–91Canada, 80–6United States of America, 76–9

senior executives, 8, 15sense-making, collaborative, 202separations of power, 91, 237shifters, 159, 160, 161, 164, 165, 166,

168, 170, 171

252 Index

Shipley, Jenny, 150, 158, 165shirking/shirkers, 159, 160, 164, 165,

169, 170, 171short-term demands versus long-term

interests, 195–6Smith, Ian Duncan, 136, 144, 149social comparison, 179social context, 178, 181social creativity, 179, 187social division, 177, 178social identity, 174

leadership change as, 175–8leadership transition and succession

and, 181–8power consolidation and leadership

change and, 178–81social identity theory, 12, 238Solicitor-General, 97, 100solidarity, 178, 179, 180, 181–2, 184

politics of, 181, 184post-succession, 186rhetoric of, 185

spill motion, 141, 142, 154n3starting a new role, 191–2stress response and/or fight-flight

response, 193strong leadership, 56, 64

see also leadershipsuccessions, 237–9

accountability perspectives on, 243–4

characteristics and challenges, 5–6conflicted, 126–7leadership, 232–4outcomes, 128–31

aspirants and new leaders, 116–21assessment, 112–16incumbents, 121–4succession process, 125–7

power perspectives on, 242–3social-identify frameworks for, 235studies, 7–9understanding, 12–17within political parties, 234–5

successor and predecessor, conflictbetween, 120–1

superordinate identity, 176, 177, 183

Talboys, Brian, 166, 169term limits, 5

Thatcher, Margaret, 96, 124, 127, 135,136, 154n3, 160, 162, 164

learning to lead in opposition, 71nationalism, 67opposition leadership and, 57, 60political learning in opposition

influence, 63–6transition to government, managing,

46Third Way, 62Timor Gap Treaty, 31transactional leadership, 57, 62–3

see also leadershiptransformational leadership, 57

see also leadershiptransition, 94, 236–7

characteristics and challenges, 2–5constitutional perspectives on,

239–40contested see contested transitionsdemocratic, 3–4, 9–10elite, 231–2executive government perspectives

on, 240–2leadership perspectives on, 11, 240legislative dimension of, 73–4peaceful transition, 1, 2political transitions, 1studies, 7–9understanding, 9–12

Trudeau, Pierre, 83, 124trust, 12, 17

through authentic leadership,building, 196–9

Turia, Turiana, 165, 167, 172n6Turnball, Malcolm, 150, 175, 186two-party system, 99, 114

United Future, New Zealand, 140, 167United Kingdom, 13, 16, 134, 136, 152,

158, 160–4, 174United Nations Educational, Scientific

and Cultural Organization(UNESCO)

Gough Whitlam and, 224–5United States of America, 175

divided government, 77–8leonine ascendancy in, 28–31nominations and confirmations,

76–7Senate, controlling, 78–9

Index 253

University of South Australia, 224upper houses, role of opposition

in, 61US economic Armageddon, 30

vulpine elites, 33–5see also elite(s)

Watson, Don, 218–19Watson, John Christian

as chairman of directors of AMPOL,216

as chairman of New South WalesTraffic Advisory Committee, 216

as director of Yellow Cabs AustraliaLtd, 216

income, 215–16influence of, 213National Roads and Motorists’

Association and, 216poor health, 212

we-ness, 238Westminster, 11, 17

cases, 133authority, for leader removal,

143–6fixed leadership terms, 138–40leader exits, reasons for, 134–6party leader accountability, in

personalized politics era, 153–4party leader security and

accountability, 146–53removal of party leaders, possibility

of, 136–7scheduled confidence votes, 140–1spontaneous leadership challenges,

141–2conventions, 95–8

norms of bureaucratic impartialityand professionalism, 241–2

opposition leadership, 58systems, 94, 114

Whitelaw, William, 87, 160, 162Whitlam, Gough, 38, 87, 95, 121,

151, 210as active leader of Caucus, 39–40as Ambassador to UNESCO, 224–5approach to preparing for

government, 38, 51as chairman of the Australia-China

Council, 224as chairman of the National Gallery

of Australia, 224failures of coordination and

consultation, 40Labor and quality of government, 41as member of 1995 Sydney Olympic

bid team, 224as member of the Constitutional

Commission, 224moving on and filling the void, 223,

224as opposition leader, 38opposition leadership and, 57, 59, 60policy development, 38–9resignation of, 41self-justification, 222transition to government,

managing, 40–1visiting fellow at the Australian

National University, 224visiting professor at the University

of Adelaide, 224Wilenski, Peter, 38–9

policy development, 38–9Wilenski, Peter, 97

254 Index

how power changes hands: transition and succession in government (transforming government)

Documents