Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Howard and Blair: comparative predominance and 'institution stretch'.

Download PDFDownload PDF

Howard and Blair:

Comparative Predominance and ‘Institution Stretch’

Conference Paper

‘Howard’s Decade’ ANU Canberra 3 March 2006

Mark Bennister Department of Politics University of Sussex Falmer Brighton BN1 9RH

+44 (0)20 8449 8566 (h) +44 (0)7801 491 928 (m)

Mark Bennister


Abstract 1

It has recently been argued that the UK premier enjoys a level of executive power unavailable to US presidents, but how does he or she compare to another prime minister operating within a broadly similar system? Commonalities of intra-executive influence and capacity exist under both premierships. Discrete institutional constraints and deviations are evident, but trends and similarities in resource capacity can be clearly identified. These include: the growth of the leaders’ office, broadening and centralising of policy advice and media operations, strengthening of the role and function of ministerial advisers. I contend that this amounts to ‘institution stretch’, with new structures, processes and practices becoming embedded in the political system by the incumbents.

Prime ministerial predominance derives from the judicious use of personal (public stature and standing, extra-executive projection, political leadership) and institutional power. This paper concentrates on the institutional and structural resources available to the prime minister and provides a timely comparative analysis of capacity within the Blair and Howard administrations, both approaching ten years in office.

In this paper I consider four key institutional resources. First, leadership resources: (patronage, prerogative and the ability to manage and set the agenda of government), second, structural resources (the strengthening and control of the centre), third, cabinet itself (management of the cabinet and its committee system provides the platform for executive control) and fourth, the centralisation of news media resources and party control.

What power have you got? Where did you get it from? In whose interests do you exercise it? To whom are you accountable? How do we get rid of you? Tony Benn 1993

1 I would like to thank Tim Bale, Paul Webb and Richard Heffernan for comments and advice relating to this paper.

Mark Bennister


We still know little about what prime ministers do, how they do it and indeed why.

We know they are dependent on others. We know they are constrained by the

environment they work in. We also know that they are powerful, influential actors.

But beyond this has our understanding of prime ministerial power moved on from

the days when it was measured by weighing up cabinet government against prime

ministerial? It has, but a notion of predominance helps us to take study of prime

ministerial power a step further.

There is little question that prime ministers are the strongest and most superior of

the actors in the executive. Favourable political circumstances sustain leaders in

power (large and stable parliamentary majority, compliant cabinet, high approval

ratings), but make them fragile once the political circumstances become less

favourable (Poguntke and Webb 2005).

This paper contends that: the British prime minister enjoys more predominance

within the core executive than the Australian prime minister, but that there are

many British characteristics of prime ministerial predominance evident in the

Australian premiership. Predominance (Heffernan 2003) allows us to see beyond

the ‘smokescreen’ of presidentialism.2 Comparative analysis of the prime minister’s

formal resources demonstrates that ‘institutional stretch’ has occurred in both

countries as the incumbents embed machinery of government changes over time.

This paper sets out a framework of comparative analysis, adapting Heffernan’s

approach and placing the research within the context of the current literature, then

moves on to analyse the institutional resources currently available to the present

incumbents in the UK and Australia, exploring each of four aspects (as legal head

of government, as cabinet manager, as controlling and strengthening the centre,

as setting the media and party agenda). Finally I draw some broad conclusions.


2 See Poguntke and Webb 2005; Foley 1993, 2000 on presidentialisation and Rhodes 2005 in particular on its limitations.

Mark Bennister


The predominance of prime ministers is recognisable in the British premiership, yet

many characteristics have been identified in recent research into premierships

elsewhere (Poguntke and Webb 2005). Such comparative research is however in

its infancy, and although there are identifiable trends in comparative research the

debate still rumbles on in Britain as to how autonomous or presidential have prime

ministers become. Contrasting views point to the academic divide as Rhodes

stresses the constraints on prime ministers as dependent actors with limited

opportunities to shape or steer, only to ‘persuade’ (in Weller et al 1997, 222). For

Foley however the prime minister is not so constrained and the dynamics of

personal leadership can have an impact on the formal support structures as well as

the wider electorate: ‘the old moorings of institutionalisation have been stretched in

response to the new context of personalised public leadership’ (Foley 2000, 341).

The core executive model views the leader as a central actor dependent on others

and institutionally constrained. Foley sees the British prime minister as

presidential, able not only to shape the institutional functions and shape of

government, but also the party and public opinion. The prime minister has

therefore transcended Rhodes’ ‘humdrum’ by reaching out beyond the traditional


My comparative analysis draws from both the core executive model and the

presidentialised model, indeed many characteristics overlap: they are not mutually

exclusive. There are strong elements of the core executive model that resonate,

but not all actors are equal or have equal resources. There are also strong

elements of the presidentialisation thesis that can be applied to contemporary

prime ministers in particular spatial leadership characteristics or the concept of

leadership ‘stretch’ whereby the influence and authority of a prime minister is

beyond the systemic. Heffernan’s personal and institutional characteristics of a

‘predominant’ prime minister have set an analysis which draws much from these

two models.

The notion of presidentialisation, then, prompts us to account for the prime minister within the core executive model. Government may not be as centralised as traditional approaches argued, but is not as decentralised as

Mark Bennister


some suggest. Well-resourced executive actors, such as the prime minister, are ‘less dependent’ on other actors, but they are never ‘totally independent’ of them (Heffernan 2005b).


Prime ministerial predominance enables the prime minister to lead, if not

command, the core executive and direct if not control its policy development.

Predominance arises from the prime minister’s ability to access and utilise

personal and institutional power resources (Heffernan 2003). Predominance helps

us to distinguish between theories of dependency (Rhodes and Dunleavy 1995;

Rhodes 2000; Richards and Smith 2002; Smith 1999, 2003) and theories of

presidentialisation (Foley 1993, 2000; Poguntke and Webb 2005).

Predominance grants the prime minister the ‘potential’ for leadership within the government, but only when personal power resources are married with institutional power resources, and when the prime minister is able to use both wisely and well (Heffernan 2003, 350).

Predominance therefore derives from this judicious use of personal and

institutional power. Heffernan (2003) identifies personal power resources as:

reputation, skill and ability; association with actual or anticipated political success;

public popularity; and high standing in his or her party. Institutional power

resources Heffernan characterises as: being the legal head of the government;

agenda setting through leadership of the cabinet and cabinet committee system

and Whitehall; strengthening Downing Street and the Cabinet Office (the centre);

agenda setting through news media management.

The combination and use of power resources is therefore an essential part of

understanding contemporary prime ministerial predominance. It locates the

research within acknowledged core executive studies and draws on newer

concepts of presidentialisation. This provides a framework for understanding what

predominance is, and how to recognise it. It also provides a comparative basis for

understanding prime ministerial predominance in two countries.

This paper will focus on the institutional (or formal) resources in the two cases

comparing and contrasting institutional power resources. We can see from this

Mark Bennister


comparative analysis that institutions have been stretched by the incumbents over

time, to the extent that capacity has increased for the contemporary and

established prime minister in the UK and Australia 3.

Prime minister as legal head of the government

The British prime minister is ‘universally reckoned to be the most powerful single

individual in the British system of government’ (King 1985, 1), often more

‘authoritative than any President‘ (Heffernan 2005, 69). The Australian prime

minister has tremendous authority merely by holding office (Jaensch 1997). The

two have different constraints (federalism, powerful Australian Senate, size of

parliament, differing modes of leadership election and so on). This paper is

however concerned more with the institutional commonalities between the two

cases: constitutional arrangements (written or unwritten), leave much to tradition

and convention, cabinet contains only parliamentarians, and the prime minister is

free to set the political agenda (Weller 1985).

The prime ministership is a ‘highly dynamic office’ (King 1985, 6), it is flexible and

depends on agency (the incumbent prime minister) and structural (institutional

constraints) factors. Power varies from prime minister to prime minister and

fluctuates within a prime minister’s period of office (Smith 1999, 73). Indeed, clarity

regarding the role and functions of Australian prime ministers is as hard to find as

in Britain.

The role of the Australian Prime Minister has sometimes been seen as merely that of a chairman of a committee. At other times that role has been seen as a presidential one (Lucy 1993, 138).

The existence of a relatively recent written Australian constitution does not assist,

as the prime minister is not mentioned, and therefore is unable to rely on it for any

formal power (Lucy 1993; Weller 2003). The key to power and predominance, as

3 The second part of the power equation, the personal, lies beyond the scope of this paper. Further research will contend that personalised leadership characteristics, extra-executive projection, and public stature transcend the institutional in both cases.

Mark Bennister


with the British premiership, lies in the discharge of informal resources, and the

management of dependency relationships alongside formal structural resources.

The leadership resources available to a prime minister (nominally residing a

Westminster system of government) tend, in the absence of any statutory

obligation, to be related to prerogative, patronage, and the power to set the

government’s agenda.

The Prime Minister's roles as Head of Her Majesty's Government, her principal adviser and as Chairman of the Cabinet are not defined in legislation. These roles, including the exercise of powers under the royal prerogative, have evolved over many years, drawing on convention and usage, and it is not possible precisely to define them (House of Commons Debates 2001).

As legal head of the government the prime minister is able to make use of royal

prerogatives 4. By convention and usage over time these formal powers can be

more precisely identified as including the power to: appoint and dismiss ministers;

summon, prorogue and dissolve parliament; appoint and regulate the civil service;

allocate and reallocate portfolios; regulate government business; create cabinet

committee; reorganise central government; confer honours; make treaties; declare

war; deploy armed forces on operations overseas (Heffernan 2005, 33; Smith

1999, 75; and Smith 2003, 62). Without any formal constitutional constraints the

British prime minister is free to exercise these prerogatives without parliamentary


However the war on Iraq sparked a debate on usage of the royal prerogative. Blair

was not obliged to take a parliamentary vote on sending troops to Iraq, but chose

to voluntarily place a substantive motion before the House of Commons Leaving

aside the various political reasons surrounding this decision, the result of the vote

was only advisory as the royal prerogative could still have been exercised. Debate

has since focused on explicitly removing the declaration of war and deployment of

armed forces from the royal prerogative and vesting the powers in parliament ( The

4 See House of Commons Public Affairs Select Committee Report 2003 on the Royal Prerogative.

Mark Bennister


Guardian 23 August 2005; House of Commons Debates 21 October 2005). This

debate highlighted the extent to which monarchical power (particularly the ability to

declare war) had passed (almost unnoticed) to the prime minister.

Similarly in Australia the prime minister’s activities are based on practice,

convention and choice (Weller 1992). The prime minister cannot rely on the

constitution for any power that other ministers lack (Lucy 1993, 138). The

constitution technically gives power to the Governor General to appoint ministers

and dissolve both houses, however in practice this falls to the prime minister to

advise the Governor General. 5

The question of prerogative also came to a head in Australia in relation to the case

of Governor General Archbishop Hollingsworth. Howard chose not to exercise his

power to advise the Queen that the Governor General should be removed from

office, following a report which found Hollingsworth had allowed a known

paedophile to continue working as an Anglican priest in Brisbane. The subsequent

furore (and inevitable departure of Hollingsworth) raised questions relating to the

prime minister’s power to recommend Governor General appointments to the

Queen, and whether this was an appropriate method of selection for the position

(Williams 2003).

The power of patronage, it is argued, helps a prime minister to cement his or her

position and to bring ministers or others into line on issues of policy (Weller 1985).

The power to appoint is wide and vast, dependency relations are very much the

result of the power of the prime minister to appoint and dismiss ministers, advisers,

committees, commissions and so on.6 The exercise of this power by Blair has been

routinely attacked and ridiculed by a personality obsessed media (with cries of

5 Section 61 of the constitution: ‘the executive power of the Commonwealth is vested in the Queen and is exercisable by the Governor-General as the Queen's representative, and extends to the execution and maintenance of this Constitution, and of the laws of the Commonwealth.’ 6

See Benn (in King 1985, 228) for his numerical estimate of patronage power of the British prime minister from 1945 to 1976 in respect of ministerial, non-ministerial appointments, creation of peers and other patronage awards.

Mark Bennister


‘Tony’s cronies’). Similarly Howard, using his patronage powers to strengthen and

support his position and has ‘pursued a highly personalised approach to

government appointments, particularly departmental secretaries and diplomatic

appointments’ (Tiernan 2006, 25). He too has been accused by his political

opponents of providing ‘jobs for the boys and girls’.7

Although ill-defined and often latent, prerogatives and conventions place the prime

minister in a ‘structurally advantageous’ position within the core executive (Smith

1999). Placed at the very top of the most important institutional hierarchy

(Heffernan 2003), the prime minister can shape the government with legal authority

derived from the crown

Controlling and strengthening the centre

Prime ministers require structural support, of a bureaucratic and a political nature.

The growth of administrative and policy capacity directly answerable to the prime

minister has been evident under Blair and Howard. Patronage enables a prime

minister to surround him or herself with key political confidants, creating an

institution of gatekeeping and dependency networks. The trick is then to establish

and embed this institution over time.

Blair transferred his team in opposition into government creating powerful advisory

positions for Alistair Campbell (Chief Press Secretary), Jonathan Powell (Chief of

Staff), Sally Morgan (Political Secretary), Angie Hunter (Special Assistant) and

David Milliband (Head of Policy) (Kavanagh and Seldon 2000). He has since

created strong advisory positions in foreign affairs, strategic planning, and delivery,

responsible directly to himself.8 The focus under Howard has less been on

powerful individuals (like Campbell, Powell, or Mandelson) behind the prime

7 The ALP unsurprisingly used the ‘jobs for the boys and girls’ claim as a campaigning tool in the 2004 election, suggesting that ‘the Howard Government’s record of more than 120 appointments over eight and a half years is well beyond acceptable community standards’ (ALP 2004). 8

By 2003 some 27 advisers out of 81 who worked across all central government ministeries in Whitehall were located in the Prime Minister’s Office (Committee on Standards in Public Life 2003: in Heffernan and Webb 2005).

Mark Bennister


minister, but on a coterie (or ‘kitchen cabinet’) of core confidents. Key individuals

important to Howard have been Lynton Crosby (Director Liberal Party), Arthur

Sinodinos (Chief of Staff), Tony Nutt (Private Secretary), and Tony O’Leary (Chief

Press Officer). These individuals span most of Howard’s period of office and have

been an important source of continuity and stability: a ring of advisers, all with the

ear of the prime minister, protecting and defending their master.

The institutionalisation of this type of policy advice of can be seen in the

establishment of small policy units close the prime minister. Blair’s policy advice

units have been fluid, often disjointed and fragmented. Machinery of government

changes to the centre have been well documented (Burch and Holliday 2004;

Hennessy 2000; Kavanagh and Seldon 2000), as producing a department of prime

minister in all but name.9 In his third term, the centre now looks a little more

coherent with a single strategy unit providing longer term policy advice to augment

the policy directorate in Downing Street. Units and advisers still straddle the

Cabinet Office and the Prime Minister’s Office, but the arrangements appear more

embedded and policy advice looks more institutionalised without any need to

create a formal prime ministerial department. The fluid nature suits Blair’s style

way of operating in informal, small groups. However the exposure of this loose and

unstructured style during the Hutton and Butler inquiries has led to a tightening up

of the processes at the core. Blair has been staunchly unapologetic about his

commitment to a strong centre, and his emphasis on bilateral meetings with

ministers.10 This bilateralism reinforces the location of the prime minister at the

centre of the core executive networks (Heffernan 2003, 360). These spheres of

contact with the dependent actors are well illustrated by Hennessy’s circles of

influence (see Hennessy 2000, 495-500).

9 ‘ There are now 190 staff in the Prime Minister’s Office, compared to 130 in 1997, and more units in the Centre as a whole. At the same time, the Prime Minister’s Office and Cabinet Office are more integrated and focused than before, with more staff working to the Prime Minister. The overall outcome is clearer lines of command and direction, and a strengthening of the position of the PM and his aides’ (Burch and Holliday 2004, 12). 10

See Liaison Committee 2002. In his first 25 months in office Tony Blair held a total of 783 meetings with individual ministers; over the same period John Major held 272 such sessions’ (Kavanagh and Seldon 1999, 275)..

Mark Bennister


In Australia the existence of a Department of Prime Minister and Cabinet (PM&C),

established in 1971 and now expanded considerably (Yeend 1979; Weller 1993;

Jaensch 1997), has of course negated the need to create even a virtual one! The

head of PM&C (most notably Max Moore-Wilton, prior to the present Departmental

Secretary Peter Shergold) has proved to be a key figure, not only driving forward

an energetic Pubic Service reform agenda, but providing key bureaucratic support

to Howard (see Davis and Rhodes in Keating et al 2000). Greater resources

supporting the prime minister do not however necessarily mean better or more

effective institutional capacity. Howard felt the need to establish a Cabinet Policy

Unit (CPU) located firmly in the core of the core executive, soon after his election

in 1996. This small unit has performed a pivotal role for Howard during his

premiership, often acting as a key power broker between the prime minister and

departments and as a key gatekeeper.11 The first head of the unit was close

Howard confident, Michael L’Estrange who also assumed the role of Cabinet

Secretary. Essentially the Cabinet Policy Unit is a political unit placed at the heart

of government, outside the jurisdiction of the Australian Public Service but under

the Members of Parliament (Staff) Act 1998. Staff are appointed by the prime

minister and accountable directly to him (PM&C Annual Report 2000-01). Under its

second head, Paul McClintock, the unit took on a strategic role driving forward

whole-of-government initiatives and playing a greater coordinating role (Howard


The focus on whole-of-government policy and related delivery of centralised

objectives are common themes to both governments. Institutionally this has seen

the establishment of the Cabinet Implementation Unit under Howard as a direct

comparator to Blair’s Delivery Unit. Howard recently spoke of the new unit:

The introduction of a Cabinet Implementation Unit which is designed in a systematic way to ensure that decisions once taken with great fanfare are not then forgotten and lose their lustre through lack of vigorous detailed implementation, so far has proved to be a valuable addition to my understanding of progress, and also that of Ministers and I think the


Private information, confidential interview 20 January 2006.

Mark Bennister


initiative which has worked well, is one that is certainly here to stay (Howard 2005).

In spite of having a dedicated department at his disposal, Howard’s own office has

continued to grow in size, following the trend set by his predecessors 12 . James

Walter wrote in the 1980s was that the institutionalisation of ministerial staffing was

'serving as another mechanism to assure prime ministerial preeminence', and the

continued growth of the office since that time lends weight to this assessment.

(Holland 2002, 10). Staff numbers in Howard’s office had grown to forty by October

2005. Of these twenty-seven are classified as advisers.13

The focus on process, exemplified by Blair’s ‘command and control’ structure at

the centre (Hennessy 2000) which peaked in 2002 (Heffernan 2005b), can be

matched by the focus on Howard’s use of ministerial advisers. With a weak

opposition and a distant parliament the Howard Ministry has accelerated the role of

advisers providing a wedge between the inner cabinet, the public service and the

legislature (this dislocation whereby ministers and the prime minister receive

politicised advice has characterised the Howard government and came to a head

in the Children Overboard Affair (Weller 2002; Keating 2003). Weller (2002) termed

ministerial advisers the politically dispensable, ‘junkyard dogs’ of the political

system (Walter 2004).

An important structural point of difference has Australian ministers based, not in

their department like British counterparts, but in their own parliamentary offices.

This produces a clear divide between the political office (containing advisers,

department liaison officers, and personal staff) and the bureaucracy. British

ministers can be regarded as ‘captive’ in their departments. While both employ an


Whitlam's office employed 21 staff, while Fraser's office had 23, despite his overall reduction in the number of ministerial staff. Hawke's office ranged from 16 shortly after the election that brought Labor to power, up to 24 by 1990. Under Keating the office had 30 staff, while under Howard it grew to 37, although Howard's ministry had only a marginally larger staff profile overall (Holland 2002). 13

Of these 40.3 staff Howard has 2 principal advisers, 3 senior advisers (PM), 7 senior advisers (Cabinet), 1 media adviser (Cabinet), 7 advisers, 7 assistant advisers, 2 personal secretaries, 6 EAOM, 5.3 secretaries (Senate Estimates 1 October 2005). For a detailed analysis of the development of the Prime Minister’s Office see Weller 1985, 139.

Mark Bennister


increasing number of ministerial staff (see Tiernan 2005) location matters, as

Australian departmental secretaries need to actively visit their ministers, while UK

permanent secretaries are close at hand. Of course depending on your view in

Australia distance from the minister can either enhance civil service ability to give

‘frank and fearless’ advice or put ministers at the mercy of politicised advice. The

lack of equivalent collegiality in the UK may be partly subscribed to the tradition of

departmentalism that locates UK ministers firmly within their portfolio silos.

Managing collegiate resources: the cabinet

The weaknesses of cabinet are well established: too little time, too much

information, too many busy people (Kavanagh and Seldon 2000, 321). Measured

by frequency of meetings and papers received, the cabinet as a set of

arrangements (Weller 2003) has steady declined over time in the UK.14 Former

Cabinet Secretary Lord Butler described Blair’s approach to the cabinet as

reverting to the 18 th century, when advisers would group around the monarch.

Cabinet under Blair is a short, informal meeting to discuss the business of the day,

involving stock-takes and lasts no more than forty minutes (Butler 2004b). 15

Whether a weberian ‘ideal type’ of cabinet is desirable is open to question: ‘If the

principles of cabinet government were applied to the letter, the system would not

merely be grossly inefficient, but truly not viable’ (Blondel in Andeweg 1997, 64).

Under Blair it is now essentially a political tool for reducing conflict, and ensuring

stability and cohesion within a single party setting.

Cabinet as defined narrowly as the weekly meeting of ministers can be both a

source of strength and weakness for a prime minister. If debate, discussion and


During the late 1940s cabinet met for an average of 87 times a year with 340 papers being formally circulated in the 1970s, 60 times a year with 140 papers and by the 1990s no more than 40 times a year with only 20 papers (Lord Butler in the Times, 22 February 1999 in Heffernan 2003, 359). This trend has continued under Blair. From 1990 to 1997 John Major chaired 271 cabinets and 189 cabinet committees and had 911 recorded meetings with individual ministers. In his first two years Blair chaired 86 cabinets and 178 cabinet committees and had 783 meetings with individual ministers (Kavanagh and Seldon 2000, 286). 15

Michael Cockerell’s (2001) illuminating BBC documentary Cabinet Confidential exposed the style of Blair’s cabinet management.

Mark Bennister


decision-making are absent, political battles will necessarily be played out in other

forums and the propensity for disputes to spill out into the public arena is

enhanced. Recent cabinet disagreements over a full smoking ban ( Guardian 27

October 2005), and education policy ( Sunday Telegraph 18 December 2005) have

exposed Blair’s leadership to even greater scrutiny. With cabinet a product of the

prime minister’s style, the announcement that he will leave office before the next

election is assumed to have eroded his authority. 16

Beyond the full cabinet there is however a vast and sprawling system of networks,

committees and taskforces where most work is undertaken. Outside formal cabinet

meetings Blair has an extensive committee structure which he dips in and out of,

chairing fourteen of them himself.17 In May 2005 Blair announced that the number

of Committees would be streamlined (reducing the total from 61 to 44) with the

expressed desire to give the committee system greater centrality. 18 Cabinet

committees enable non-cabinet ministers and officials to be involved in policy

discussions: ‘a passport to involvement’ (Dunleavy 2003, 344). However, clearly

some have much more importance than others. Dunleavy’s study (2003) of cabinet

committees showed the extent to which Blair conceded a greater role on domestic

affairs to his Chancellor than his predecessor (ibid, 351). The more hands-on

approach by Blair to committees after the 2005 election redressed this somewhat

(although it is evident that ‘policy fiefdoms’ are clearly divided between Blair and

Brown (Hennessy 2005, 10)). It remains the case that the bulk of Blair and Brown’s

duopoly business (ibid) is conducted through bilateral meetings. Any partial return

to the use of Cabinet committees has been subscribed to Prescott’s keenness on

them (ibid). Since the Hutton Inquiry exposed the ad hoc nature of Blair’s informal

style of government (the so called ‘sofa approach’), with little formality and no

recorded minutes, there has been much more minute taking (Butler 2004a;

Hennessy 2005). Despite the institutionally complex system of formal committees


Blair gave an interview to the BBC's Andrew Marr, on the eve of the Labour Party’s 2004 Annual Conference, in which he stated that he would serve a full third term if elected, but then stand down ( accessed 16 Jan 2005). 17

See Catterall and Brady in Rhodes 2000 on the development of Cabinet Committees in Britain. 18 Subsequently two more were added in Dec 2005 (PM press release 15 Dec 2005).

Mark Bennister


at his disposal, Blair is still keen on the informal.19 Avoiding internal cabinet fallout

and the power struggles that dogged the Major government has always been a

priority, yet this has served to stifle and homogenise collegiality in his largely

subservient cabinet.

Cabinet in Australia reflects the more institutional and collegiate approach evident

in Australian political culture. Big issues (such as military action in Iraq) are

discussed at length and over time and rugged debate is expected (Weller 2003). 20

Howard is a cabinet traditionalist (Tiernan 2006) and likes to use cabinet meetings

as a sounding board, to test the public line on salient issues of the day, act as a

pressure valve, and to bind colleagues into the party line.21 Until procedures were

streamlined in April 2002, allowing more time for strategic discussions, formal

submissions to Cabinet were long, detailed and tended to occupy most of the

discussion time within cabinet.22

Cabinet support for Howard is organised, formal and important. It remains at the

central apex of government. It has been streamlined and the committee structure is

lean and focused.23 Howard himself chairs four of the five cabinet committees

including the key Expenditure Review Committee (ERC). Howard’s role in budget

formation is interesting to contrast with Blair’s minimal role in financial matters.

According to one senior source the difficulties Blair faced in cabinet over Iraq

(whereby cabinet differences were played out publicly) and the decision to have a


Blair is on record as saying he expects to hear of any minister’s concerns well before formal meetings (Liaison Committee 2002; Cockerell 2001). 20 PM&C Annual reports the cabinet secretariat supported 57 cabinet meetings (including NSC and committee meetings) in 04-05. This compares with 120 in 97-98 and the high point of 141 in 99-00. Since the streamlining in 01-02 of cabinet submission which reduced full cabinet handling of many submissions, meetings have stayed constant at between 57 and 67 per year (PM&C Annual Reports). 21

Private information, confidential interview 7 January 2004. 22 Private information, confidential interview 7 January 2004; see Howard 2002 background notes on cabinet procedures regarding move towards more a more strategic cabinet system. 23

‘His [Howard’s] new blueprint for government and especially cabinet is not likely to be ditched by his successors ‐ Labor or Liberal ‐ because it works too well. It minimises the opportunities for damaging differences, time consuming discussions between ministers, and leaks’ ( Sydney Morning Herald 16 August 2005).

Mark Bennister


referendum on the EU Constitution (Jack Straw admitted that the decision to have

a referendum overturning previous policy had not been discussed formally in

cabinet ( The Guardian 20 April 2004)) could not happen under Howard.24 The

National Security Committee (NSC) has grown to rival the ERC as the most

important cabinet committee. Howard regards the NSC (comprising of the six most

senior ministers and the key agency heads) as ‘one of the very significant

successes’ of his government in terms of governance arrangements (Howard

2005). Others have suggested that the NSC has been part of the trend of power

centralising around the prime minister, consolidating his position as the unrivalled

source of power and authority for national security policy making. 25 This cabinet

system looks more like a collegial system, one where ministers are involved and

feel ownership of the process. Leaks are rare (although those that have occurred

inevitably focused on the Howard-Costello relationship) and differences played out

behind closed doors. 26 The institution has been stretched to provide Howard with

the authority of cabinet collectivity, new centres of power have arisen within the

network (NSC and CPU) to increase capacity within the structure.

Importantly, though Howard must work with a coalition partner in cabinet, and while

the Nationals and Liberals have generally had a trouble-free relationship, recent

events (sparked by the defection of Senator McGauran to the Liberals) show the

potential for dislocation between the two parties. In addition, in choosing his

cabinet, Howard has needed to balance the usual political demands with those of

state representation. (The disproportionate number of Scottish cabinet ministers

over time suggests that regional balance has not been an issue for Blair!)

This power to appoint cabinet ministers is again a source of strength and

weakness. The power of Brown to set the tone for all ministerial appointments to

be considered as either Brownite or Blairite demonstrates the level of duopoly in

the relationship. Blair may have the formal power to appoint, but he has a strong


Private Information, confidential interview 10 May 2004. 25 Peter Jennings, Australia Strategic Policy Institute quoted in The Australian ‘More Power to the PM’ 29 October 2005. 26

See Australian Financial Review 6 June 2005 on Singapore Airlines.

Mark Bennister


political constraint in the shape of his powerful chancellor. His reshuffles over the

years have proved less than successful (the worst example being the messy June

2003 reshuffle that abolished the post of Lord Chancellor and gave birth to the

Department for Constitutional Affairs), as he has tried to instigate machinery of

government changes at the same time as balancing personnel matters. 27 Blair

has lost key confidants from his cabinet (Mandelson twice, Blunkett twice, Byers,

and Milburn), while Brown has remained along with Prescott (as the link to the

party) and Beckett as the only survivors from his first cabinet.

Howard has similarly maintained his faith in his Treasurer Costello as the heir

apparent. The other constant cabinet minister has been Downer (who as the other

heavyweight cabinet minister has remained as Foreign Minister for the duration).28

Howard’s reshuffles have been fewer in number and more considered than Blair’s

(the most recent reshuffle sparking tension with the Nationals is an unusual

departure for Howard). They have tended not to be accompanied by large

machinery of government changes. Of course there is a smaller pool of aspirants

to choose from, and mindful of the early resignations after the travel rorts affair

Howard has shown a marked reluctance to lose ministerial colleagues during his


Agenda-setting through the news media and party control

For Hennessy (2000, 483) and others (Riddell 2000) the intersection between

media management and Whitehall came with the publication of the new Ministerial

Code Of Conduct in 1997 and in particular the contentious paragraph 88:

In order to ensure the effective presentation of government policy, all major interviews and media appearances, both print and broadcast, should be agreed with the No 10 Press Office before any commitments are entered into. The policy content of all major speeches, press releases and new policy initiatives should be cleared in good time with the No10


Seldon 2005 notes that Blair tends to leave reshuffle decisions to the last minute and consult only a very small number of trusted advisers 28

Senator Vanstone started in Howard’s first cabinet and after being dropped, returned to a key portfolio. Key Senate fixer Robert Hill recently stepped down. 29

His ‘aversion to the removal of ministers’ is said to relate to the 1997 travel rorts scandal, when five frontbenchers and two staffers resigned ( Age 2005).

Mark Bennister


Private Office, the timing and form of announcements should be cleared with the No 10 Press Office 30 .

For many traditionalists, like Riddell, this represented ‘the biggest centralisation of

power seen in Whitehall in peacetime’ (Hennessy 2000, 484). Arguably, however it

was merely the transference of the strict command and control system of media

management that had been conceived and put into practice in Opposition by New

Labour. Structural changes were swiftly implemented, with the creation of the

Strategic Communications Unit (SCU) in Downing Street, accountable to the prime

minister through the Chief Press Secretary, Alistair Campbell. The SCU replicated

the Millbank system of strong central coordination and key headline messages that

the New Labour elite had been familiar with in opposition (Kavanagh and Seldon

2000, 255). The focus on media relations translated into a big rise in the number

staff employed in the Press Office and SCU (Hennessy 2000, 485). The operations

were designed to be highly efficient in both reactive and proactive relations with the

media, being able to respond and set the agenda from the centre.31 Although the

obsession with ‘spin’ may have lessened over time and post Hutton, media

management has now stretched the institutional mechanisms to the extent that

communication strategy is an embedded function of the Blair government.

Institutionalised media management capacity is not as extensive and pervasive

under Howard. However the media unit in the PMO is ‘the largest ever assembled

by an Australian prime minister’. 32 Howard’s own Alistair Campbell, Tony McNulty

centrally coordinates the activities of thirty-four media staff serving the Ministry. Yet

a comparison with the media management and response capacity developed

under Blair, needs to predate the Howard era. The National Media Liaison Service

(NMLS), established in 1983 under the Hawke administration and then providing

support to the Keating premiership, can be seen as the blueprint for the New


The Ministerial Code replaced the Questions of Ministerial Procedure. See Baker 2000 for a full discussion of the genesis of the Code and the code itself. 31 For detailed information on the process see Kavanagh and Seldon 2000, 254-261, particular details of ‘the Grid’. 32

The Media Unit has a staff of eight, comprising a press secretary, a senior media advisor, a media advisor, an assistant media adviser, and four media assistants (Tiernan 2006,17).

Mark Bennister


Labour approach to centralised media operations. It monitored media from all

sources around Australia and produced detailed briefings for ministers and

parliamentarians efficiently and effectively (Holland 2002). Howard and the Liberals

campaigned against the NMLS in opposition and disbanded it on entering office,

outsourcing media monitoring (Barns 2004). The need for a similar coordinated

and centrally directed unit has seen the Government Members Secretariat (GMS)

grow in prominence under Howard. The unit channels government messages to

parliamentarians reinforcing discipline and coherence. This helps to bind the

parliamentary party together with the executive. Under Howard, the party in power

is regarded as the government. With a smaller parliamentary party, discipline is

important and with disputes confined to the party room, backbenchers have an

important role in selling government policy. 33 The machinery is limited in its reach

to the states, although the administration has placed a media officer in each of six

ministerial offices, one in each state to coordinate and disseminate whole-of-government strategy (Holland 2002, 33).

Prime ministers in the UK and Australia are both products of their parties. The

relationship with the party is crucial (Poguntke and Webb 2005) to understanding

the ability and capacity of prime ministers to ‘stretch institutions’ and manage

limitations and political constraints. Blair’s party reforms established New Labour

as an efficient, uncompromising election winning machine. The New Labour project

was conceived in opposition and borne of three successive election defeats.

Discipline, subjugation to a charismatic leader, and constant adherence to the

central message characterised the party reforms. It placed the two protagonists

Blair and Brown in unassailable charge of the party in the pursuit and maintenance

of power in office. 34 Blair has little direct accountability to the Labour party.

Meetings of the backbench Parliamentary Labour Party are generally treated with

disdain, and Blair has had little time for traditional Labour party structures (annual

conference has been downgraded, decisions are routinely ignored and the ruling

executive committee neutered). The consequence of this approach was supposed


Private Information, confidential interview 20 January 2006. 34 See Rawnsley (2001) and Naughtie (2001) for journalistic accounts of the Blair-Brown relationship.

Mark Bennister


to lead to a greater strengthening of the leadership, and an elimination of the

possibility of a Labour government suffering at the hands of its members. However,

the weakening of party ties has dislocated Blair from the party and a falling

parliamentary majority has now made Blair more vulnerable to defeat in the House

of Commons 35 .

The comparative angle here has much to do with numbers. The Labour Party has

353 MPs currently and a majority of 65 out of a legislature of 646. The Coalition by

contrast has 74 Liberal MPs and 12 National Party MPs out of 150 member lower

house. Howard governs with a majority of 22 and until 2004 did not command a

majority in the Senate. Discipline, in Australia, is therefore very important, MPs

very rarely cross the floor to vote with the opposition. Any differences are thrashed

out in the party room, which consequently becomes an influential point of contact

for Liberal MPs. It has even been known for decisions made in Cabinet to be

overturned after frank party room discussion. The parliamentary Liberal Party as

with the ALP, has the power to elect its leader, whereas now all the main parties in

the UK involve the wider party membership in leadership elections. (Although if

Blair were unable to sustain the support of the parliamentary Labour party, he

would find it almost impossible to remain in office.) The impact of the party room

means that Howard is ‘less likely to run a one-man band’ (Weller 2004, 639).

Whilst the party is an important player, it is worth noting that Parliament is in

session for less than 100 days per year, while Cabinet is constantly in session

(Jaensch 1997,167).

Strong party discipline and a capacity to absorb the lessons of a long period in

opposition have characterised both periods of office. Yet as Blair and Howard

inevitably both approach the end of their tenures, cracks are certain to appear

amid the political jockeying for position. Importantly though the inheritors of the

mantle will be faced with enhanced prime ministerial capacity and stretched

institutions as the next prime minister builds on legacy of the predecessor.


This is primarily due to the existence of a hardcore of over twenty Labour rebels, who now attract additional malcontents on various policy issues (see Cowley 2005).

Mark Bennister



Predominance is defined by a combination of the use of personal and institutional

resources. This paper has considered the institutional capacity maintained and

generated under Blair and Howard. These formal powers have been ‘stretched’ by

the incumbents. In two political systems characterised by convention and tradition,

the capacity to push the boundaries is enhanced. Constraints exist and have

arisen over time (lack of Senate control, rebellious Labour backbenchers), but the

ability to create new institutions (policy, delivery and media units) and shape

existing ones (via patronage), with an array of advisers at the prime minister’s

disposal, has enhanced formal capacity. However incumbency, the formal

authority derived from successive electoral victories, and weak political opposition

are powerful determinants, common to both Blair and Howard.

Important points of difference do help us to understand how and why prime

ministers act as they do. Institutionally the Australian prime minister is constrained

by the federal structure and the need to negotiate with the states (which since

2002 have all been ALP controlled), accommodation of a coalition partner in

government, and accountability to the party room. However these constraints have

not encumbered Howard in constructing a powerful centre based on cabinet

cohesion, stable and concentrated advice structures, and increased personal

capacity. The existing institution of cabinet is the bedrock that has been stretched

and shaped to reinforce a less charismatic but equally unyielding leader. The

tradition of cabinet collegiality is though a part of the Australian political culture,

whereas the function of collegial government has long been in decline in Britain (if

it ever existed). It is elsewhere that we find the institutional stretch under Blair. It is

located more in the informal networks of decision-making, the fluidity of the centre,

and the power of the duopoly with Brown which has created two distinct

institutional clusters.

Within the institutional setting we can identify common trends, tighter and more

responsive advise structures, centrally located media functions, patronage and

Mark Bennister


dependency networks. All these add to prime ministerial capacity. These trends,

are not explained merely by more autocratic leadership styles and incumbency

factors. The internationalisation of politics (especially since 11 September 2001),

and the growth and complexity of the state (a fragmented state needs greater

strategic coordination from the centre) have a causal role in explaining intra-executive predominance (see Poguntke and Webb 2005). In the face of this

changed political climate, institutions have been stretched and moulded to fit a

contemporary dynamic.

The key test of how predominant these two prime minister’s have been will be the

extent to which their successors maintain the institutional stretch that has occurred,

only then will we see the true impact of the prime ministerial legacy .

Mark Bennister



Andeweg, R (1997), ‘Collegiality and Collectivity: Cabinets, Cabinet Committees and Cabinet Ministers’, in Weller, P., Bakvis, H., and Rhodes, R. A W. eds. (1997) The Hollow Crown: Countervailing Trends in Core Executives, Basingstoke: Macmillan.

ALP (2004), Machinery of Government: The Labor Approach . Canberra, Australian Labor Party Policy document. August 2004.

Baker, A (2000), Prime Ministers and the Rule Book London: Politicos.

Barns, G (2004) ‘Inside the Propaganda Machine’ Australian Policy Online 20 July 2004 accessed 21 January 2006.

Burch, M. and Holliday, I. (2004). ‘The Blair Government and the Core Executive’, Government and Opposition Vol. 39, No. 1: 1-21(21) (January 2004).

Butler, D. (1973), The Canberra Model: Essays on Australian Government, London: Macmillan.

Butler, Lord (2004a), Review of Intelligence of Weapons of Mass Destruction, Report of a Committee of Privy Counsellors , HC 898, HMSO: London 2004.

Butler, Lord (2004b), ‘Cabinet Government in the UK’, Speech Shapiro Government Society, London School of Economics, 10 February 2004.

Cockerell, M. (2001), Cabinet Confidential , audio visual 17 November 2001 BBC2.

Cowley, P (2005), The Rebels: How Blair Mislaid His Majority , London; Politicos.

Dunleavy, P. (2003), Analysing Political Power: Developments in British Politics , P. Dunleavy, A. Gamble, R. Heffernan and G. Peele, Basingstoke: Palgrave Macmillan.

Foley, M. (1993), The Rise of the British Presidency , Manchester: Manchester University Press.

Foley, M. (2000), The Blair Presidency , Manchester: Manchester University Press.

Heffernan, R. (2003). ‘Prime ministerial predominance? Core executive politics in the UK.’ British Journal of Politics and International Relations 5(3): 347-372.

Heffernan, R and Webb, P (2005) ‘The British Prime Minister: More Than First Among Equals’ in Poguntke, T and Webb, P (eds), The Presidentialization of

Mark Bennister


Politics: A Comparative Study of Modern Democracies, Oxford: Oxford University Press.

Heffernan, R. (2005a) ‘Why the Prime Minister cannot be a President: Comparing Institutional Imperatives in Britain and America’ Parliamentary Affairs Vol. 58 No.1 2005: 55-70.

Heffernan, R. (2005b) ‘Prime Ministerial Power and its Limits: Core Collegiality and the UK executive’ Paper presented at the Political Studies Association, Annual Conference, University of Leeds, 4-6 April 2005 unpublished mimeo.

Hennessy, P. (2000), The Prime Minister: The Office and Its Holders Since 1945 , London: Penguin.

Hennessy, P (2005) ‘Rules and Servants of State: The Blair Style of Government 1997-2004 Parliamentary Affairs Vol 58 (1) 2005 6-16.

Holland, I. (2002), Accountability of Ministerial Staff? Research Paper No. 19 2001-02 . Department of the Parliamentary Library.

Howard J (2002) ‘Strategic Leadership for Australia Policy Directions for a Complex World’, Speech 20 November 2002 accessed 21

January 2006.

Howard, J. (2005) Speech, Ministerial Conversations Lunch Seminar, The Great Hall, Parliament House 12 August, accessed 21 January 2006.

Jaensch, D., (1997), The Politics of Australia (2 nd edition) Melbourne: Macmillan.

Kavanagh, D., and Seldon A (2000), The Powers Behind the Prime Minister London: HarperCollins.

Keating, M. (2003). ‘In the Wake of A Certain Maritime Incident’, Australian Journal of Public Administration 62(3): 92-97.

Keating, M., Wanna J., and Weller P., eds., (2000), Institutions on the Edge?: Capacity for Governance , St Leonards: Allen & Unwin.

King, A. (1985), The British Prime Minister , Basingstoke: Macmillan.

Liaison Committee (2002), Oral evidence presented by Rt Hon Tony Blair MP, Prime Minister on 16 July 2002 , HC1095, published 26 July 2002 accessed 13 June 2005)

Mark Bennister


Lucy, R. (1993), Australian Forms of Government: Models in Dispute 2 nd edition, Melbourne: Macmillan.

Naughtie, J (2001), The Rivals: The Intimate Story of a Political Marriage , London: Fourth Estate.

PM&C 2001, Supporting the Cabinet - Special Report. Annual Report 2000/01 . Canberra, Department of Prime Minister and Cabinet.

PM&C 2002, Annual Report 2001/02 . Canberra, Department of Prime Minister and Cabinet.

PM&C 2003, Annual Report 2002/03 . Canberra, Department of Prime Minister and Cabinet.

PM&C 2004, Annual Report 2003/04 . Canberra, Department of Prime Minister and Cabinet.

Poguntke, T., and Webb, P., (2005), The Presidentialization of Politics: A Comparative Study of Modern Democracies Oxford: OUP.

Rawnsley, A. (2001), Servants of the People: The Inside Story of New Labour , London: Penguin.

Rhodes, R. A. W. and P. Dunleavy, eds. (1995). Prime Minister, Cabinet and Core Executive , Basingstoke: Macmillan.

Rhodes, R. A. W. (2000), Transforming British Government , Basingstoke: Macmillan

Rhodes, R. A. W (2005), ‘Presidents, Barons, Court Politics and Tony Blair’. Paper to the panel on ‘Interpreting British Politics’ at the Political Studies Association, Annual Conference, University of Leeds, 4-6 April 2005 unpublished mimeo.

Richards, D. and M. J. Smith (2002), Governance and Public Policy in the UK , Oxford: Oxford University Press.

Riddell, P. (2000), Parliament Under Blair , London: Politicos.

Seldon, A. (2001), The Blair Effect London: Little Brown.

Seldon, A. (2004), Blair London: Free Press.

Seldon, A and Kavanagh D (2005) The Blair Effect 2001-2005 Cambridge: Cambridge University Press.

Smith, M. J. (1999), The Core Executive in Britain , Basingstoke: Macmillan.

Mark Bennister


Smith, M. J (2003)‚ The Core Executive and the Modernisation of Central Government’ in Dunleavy, P, Gamble, A. Heffernan, R., and Peele, G., (eds) Developements in British Politics 7 Basingstoke: Palgrave Macmillan.

Tiernan, A (2005) ‘Minding the minders: addressing the problems of ministerial staff’, Australasian Political Studies Association Conference Otago University, Dunedin, New Zealand 28-30 September 2005, unpublished mimeo.

Tiernan, A (2006 forthcoming), ‘Advising Howard: interpreting changes in advisory and support structures for the Prime Minister of Australia’ Australian Journal of Political Science.

Walter, J. (2004), ’Why Prime Ministers Go Too Far: Institutional Change, Security and the Renaissance of Groupthink’. New Security Agendas: European and Australian Perspectives Conference, Kings College London, 29 September - 1 October 2004, unpublished mimeo.

Weller, P. (1985), First Among Equals , Hemel Hempstead: Allen and Unwin.

Weller, P., ed. (1992), Menzies to Keating: The Development of the Australian Prime Ministership , Melbourne: Melbourne University Press.

Weller, P. (2002), Don't Tell the Prime Minister , Carlton North: Scribe.

Weller, P. (2003), ‘Cabinet Government: An Elusive Ideal’, Public Administration 81(4): 701-722.

Weller, P., Bakvis, H., and Rhodes, R. A W. eds. (1997) The Hollow Crown: Countervailing Trends in Core Executives, Basingstoke: Macmillan.

Williams, P (2003) ‘Political Chronicles’ Australian Journal of Politics and History Volume 49 (4) 2004: 558-621.

Yeend, G. (1979) ‘The Department of Prime Minister and Cabinet in Perspective Australian Journal of Public Administration Vol. 38, No.2 (June 1979): 133-150.