Comments on “Independent Review of the APS: Priorities for
The document recognizes some undesirable consequences of the 1980s “New Public
Management” fad. These include a loss of public servants’ capacity to work on policies and
programs crossing portfolio lines, and the pursuit of performance metrics at the expense of
achievement of outputs and outcomes.
My comments, in response to the authors’ invitation to “challenge our thinking”, relate to
three other areas:
The inherent conflicts in serving “the government, the parliament, and the Australian
public”. It is possible that the traditional winner-take-all “Westminster system” with a
public service closely associated with executive government, is not fitted to the way
Australian democracy is evolving.
The need for professionalism. The document correctly asserts that all staff should be
“professional public servants” but it seems to confine the notion of “professionalism” to
the acquisition of skills, without consideration of professional responsibilities and norms
and codes guiding professional behavior.
The need for leadership that allows public servants to help Australians cope with
adaptive change. It appears that the authors have assumed that “leadership” is
synonymous with the exercise of authority.
- Serving the government, parliament and the Australian public
In the theory of principal-agency behavior as applied to democratic government, there
should be no conflict between serving all three groups – executive government, parliament
and the public. But no democracy is perfect, and the reality is that there are conflicts.
Although there is no constitutional direction specifying the role of the public service, by
convention it serves executive government, which has objectives that don’t necessarily align
with those of parliament. As a minister in the Morrison Government put it “I’ve always seen
Parliament as a disadvantage, frankly, to sitting government”.1 Possibly this was a loose
statement by a politician struggling to understand the ground rules of democracy, but a
conflict between executive government and parliament is manifest during question time
and in Senate hearings, and is also seen in conflicts over important legislation, most starkly
in recent times over medical evacuation for asylum-seekers in offshore detention.
Then there is the question of serving the Australian public. The constitutional and legal
arrangements that provide people with an elected parliament are always conditional, never
entirely settled. Important questions remain unresolved: perhaps they can never be
resolved. Are members of parliament delegates or representatives? Can the voting system
Quote from Minister from Home Affairs, Peter Dutton, videoclip on ABC 12 December 2018.
2 Ian McAuley be gamed? How does voting capture the strength of interests? Do the interests of voters
correspond to the interests of the non-enfranchised – the young, recent immigrants, future
generations? Political theory does not provide clear-cut answers to any of these questions.
In a traditional two-party “Westminster” system, however, an arrangement whereby the
public service is answerable to executive government is administratively functional, even if
it falls short of democratic principles, given the “winner-take-all” outcomes of Westminster
But that system is coming apart. Political alignments are cutting across traditional “left-
right” and “progressive-conservative” divisions that once defined partisan ideologies.
This is most clearly manifest in Australian voting patterns over the last eighty years, as
shown in the figure below. The two-party system, as revealed in voting patterns, is
unraveling, and that unraveling has been particularly strong in the last thirty years.
In terms of political representation there is a time lag, because primary votes for the main
parties have to fall below certain levels before the results show as seats won in parliament.
Those results are likely to be manifest first in the Senate, where there is proportional
representation, and in electorates with strong regional issues transcending party barriers.
Although in the postwar years the party forming executive government has rarely held a
majority in the Senate, many administrations – Coalition governments in particular – have
enjoyed the support of minor parties in the Senate to pass important legislation.
But a Senate alinged with executive government is no longer the norm, and in recent times,
including the present, the party forming executive government has not had enjoyed a
majority in the House of Representatives. Some may regard the experiences of the Gillard-
Comments on Review of the APS 3
Rudd and Morrison administrations to be statistical anomolies, resulting from elections with
close outcomes. For example the media still use the term “hung parliament”, but the trend
seems to be established. It is manifest not only in Australia but also in democracies on the
European mainland, where formal and informal coalitions are the usual outcome of
elections. In Britain, the US and Canada the trend is less established because of their first-
past-the-post voting systems, but In Britain the manifestation is in terms of a breakdown in
party discipline in the Brexit deliberations.
This means that legislation is often being amended, or increasingly initiated, in parliament.
But those politicians involved generally do not have the assistance of advice from the public
service. They do have resources such as the Parliamentary Library, and the Parliamentary
Budget Office, but these are small and overworked in comparison with the public service. At
times they may be given special and limited access to senior public servants (at the behest
of the relevant minister), but senior public servants do not have the specialist knowledge of
precedents and unintended consequences that more specialist staff down the line have.
Many parliamentarians have little knowledge of the workings of public administration.
Theoretically, in a Westminster system with alternating partisan government, the
opposition has members with ministerial experience, some of whom occupy “shadow”
portfolios, but independents and members of minority parties don’t enjoy that benefit.
There needs to be far more flexible arrangements to ensure that parliamentarians have
access to the knowledge, experience and memory of the public service. Of course that
would be inconsistent with the current arrangements whereby public servants are co-opted
into partisan service to executive government, and where public servants are often in the
role of defending executive government against parliament. (The need for a more
professional public service, serving but not subservient to executive government, is covered
in the section on professionalism).
That would give meaning to the authors’ reasonable assumption that the public service
serves not only government, but also parliament.
As the authors point out, there is declining trust in traditional institutions. They point to
dissatisfaction with public services.
But this generalization needs unpacking, because in spite of a general decline in trust, some
institutions, including public institutions, are trusted more than others. The diagram on the
next page, based on surveys by Essential Media,2 paints a fuller picture.
Essential Report 13 March 2019 Trust.
4 Ian McAuley
At the top are public institutions that are somewhat separated from executive
government – police (state and federal), the High Court, the ABC, and the Reserve Bank.
Next down is the Commonwealth public service.
Then parliament (state and federal).
And at the bottom are political parties.
Executive government is closely associated with political parties, particularly in a period
when governments seem to be in continuous election mode.
Although the public service is trusted in itself, if it is too closely identified with executive
government it is in danger of being dragged down to the same level of mistrust. The public
may not necessarily consider the public service to be partisan, but if the public see it as too
“responsive” – ready to swing in to support and reinforce the ideology and spin of whatever
party takes office – the effect of guilt by association prevails.
Executive government may find it convenient to have a loyal and “responsive” public
service. It’s a tremendous administrative asset in terms of providing secretariat services
(handling correspondence, drafting speeches) and a tremendous political asset in its
capacity to present the government’s platform in a positive light, while engaging in all the
communication devices of sophistry, obfuscation, and suppression of information (while
generally being careful not to cross the boundary that formally constitutes lying).
Such subservience is at the cost of provision of “frank and fearless advice”, as many have
pointed out. It means that the public service becomes, de-facto, an extension of the party in
Comments on Review of the APS 5
Perhaps the public may accept that access to a loyal and responsive public service is a
trophy to be enjoyed by the winning side and will be balanced in the longer term as
elections are won and lost, as if politics is similar to a football competition, where similar
teams play for similar objectives. (It’s notable that many commentators use sporting
metaphors to describe political contests.)
But although many observers have pointed to a convergence of ideologies of “left” and
“right” parties, political parties are not look-alikes wearing different colored sweaters. One
manifestation of difference is between “conservative” and “reforming” parties. In Australia,
in general, the Coalition is conservative while Labor is reforming, but there are important
exceptions, such as the Hewson “Fightback” policy in the 1990s (unsuccessful), and the
radical agenda of the Abbott Government.
A conservative government, concerned to maintain the status quo, will be well-served by a
compliant public service, even if it is tainted with mistrust by association with the
government in office. But a reforming government will be better served by a public service
that is seen as professional, and at arm’s length from government. If a government, in
pursuit of the common good, seeks to implement policies that would cause real (or even
perceived) pain or distress for some groups, then it needs the authority of sources of advice
trusted by the community. After all, a reforming government, almost inevitably, will face
opportunistic criticism from opposition parties and from some lobby groups.
In this regard it is hardly surprising that the Reserve Bank rates highly on the trust scale,
even though its decisions on interest rates (or even its absence of decisions) cause pain and
difficulties for some groups. It is not seen as the handmaiden of the party in office.
Another institution worthy of mention, although not covered in the Essential survey, is the
Productivity Commission. Its clear separation from industry and other economic
departments has been crucial in helping governments pursue reform agendas. (Scholars of
public administration in other democracies see the Productivity Commission as an
institution they would like to see established in their own countries.) Of course
governments do not necessarily accept all the recommendations of the Commission, but the
Commission does make reform easier.
Similarly the Australian Bureau of Statistics is trusted because of its uncompromising
An institution that has historically been seen to have some distance from executive
government has been the Treasury Department, particularly before it was broken into two
departments in 1976. It was once seen as a bulwark against the spending whims of
government, representing what some would refer as “the permanent will of the people”.3
Its behavior during the period of the Whitlam Government remains controversial to this
day, but the general principle of having a body responsible for preserving, or at least
overseeing, the condition of national treasure (in that case the government fiscal coffers) is
an important one, particularly in a country with a short three year election cycle. The same
John Wanna, Joanne Kelly, and John Forster, Managing Public Expenditure in Australia, Sydney: Allen and
6 Ian McAuley principle should hold for other assets that can be depleted by an opportunistic government
seeking short-term political gain. Such assets, include physical infrastructure, environmental
resources, and less tangible assets such as education and social capital. It would be
inconsistent with democratic principles (and the Australian Constitution) to give such
institutions the authority to thwart the desires of elected government, but they should be
required to report on the condition of national assets without the fear of any form of
censorship or other retribution from executive government.
There is no simple model of administration that can satisfy, on the one hand a government’s
desire for responsiveness and support, and on the other hand the need for a trusted source
of advice and the community’s desire to see preservation of national assets. But in general
there should be a greater distance established between all government agencies and
Also much can be done in the form of processes to make appointments to senior positions,
and development of a culture of professionalism in the public service.
The authors call for “formal, focussed professionalisation of all APS roles”. Importantly they
report on skills gaps and the presence of “unused potential”.
These are important issues. In its recruitment the public service does well, but it does not
use or develop that human capital to its full potential.
In evidence assembled over 16 years of teaching around 1600 mainly Canberra-based public
servants, a colleague (the late Helen Coventry) and I found many shortcomings.4 There was
little discontent with general mandated pay and conditions – “hygiene factors” to use the
term of academic management theory5 – but we found several other sources of serious
• the disheartening effect of Coalition politicians devaluing the public sector,
promulgating the idea that it is intrinsically inefficient or even a worthless overhead;
• poor service by the public sector unions, who work to an outdated industrial model;
• soft” bosses, who fail to deal with underperformers;
• skills mismatches, a lack of valuation of professional skills, and a demonstrably false
assumption that seniority and professional competence are correlated;
• contracting out of research and policy advice, ignoring “in house” skills and
experience – a source of annoyance when consultants are overpaid, when they
Helen Coventry and Ian McAuley, “A thing or two we learned from public servants in a management course”,
Working document 2017.
Frederick Herzberg, “One More Time: How do You Motivate Employees?” Harvard Business Review, January-
Comments on Review of the APS 7
produce poor quality or conflicted work, and when the ease of employing consultants
contrasts with the impediments imposed by staff ceilings;
• poor appointments of senior staff, based more on loyalty to and conformance with
agency traditions than on policy or administrative competence;
• a general management culture impeding productivity – a lack of access to resources
required for people to work efficiently, a lack of cost-awareness (manifest in public
servants being directed to unproductive work), open office work environments;
• poor quality training run by big-name consultants;
• politicization – particularly in departments under direct ministerial control where
there are perceptions of corruption and where there is co-option of public servants
for partisan work (less prevalent in other agencies such as the ABS);
• a culture that places support for executive government above any notion of public
• a general culture of “learned helplessness” manifest in people behaving with a lack of
energy, confidence or enthusiasm, and in some cases deep cynicism.
Some of these findings align with the authors’ findings about skill deficits. But they go
further, suggesting that there is a lack of professional respect in the public service. In spite
of rhetoric about “flatter structures”, the public service remains what organization theorists
call a “machine bureaucracy”, organized along hierarchical Weberian lines, rather than a
“professional bureaucracy”, with respect not only for professional standards of expertise
but also for professional standards of behavior.6
Many administrative functions of the public service are appropriately organized as machine
bureaucracies, but that model is entirely inappropriate for work involving research, policy
design, and consultation with stakeholders.
Professional standards need to incorporate not only specific qualifications, but also
responsibilities and duties. Those entrusted by the public to attend to the public purpose
should be held to high standards of honesty and integrity.
This may seem to state the obvious, but public servants are often called upon, or take it
upon themselves, to act unethically in service of their political masters. Rarely does this
extend to blatant lying, but there are many other behaviors that have the effect of
misleading or holding important information from the public, in order to provide ministers
with cover. Confidentiality is often used for no reason other than to protect politicians or
interest groups supportive of the government from embarrassment. More generally, there
are ways in which language can be and is used to mislead, such as:
• presenting numerical data using the most favorable frame possible, deciding whether
or not to normalize for inflation or population growth, choosing convenient start
dates for time series and so on;
Henry Mintzberg, “Organization design: Fashion or fit”, Harvard Business Review January 1981.
8 Ian McAuley
• using sophistry to frame statements in ways that are logically correct but misleading
– “this decision was taken after consideration of scientific advice”, “we cannot
guarantee that there were no terrorists on that boat”;
• using vague language, ruling out the possibility of verification – “it is generally
believed that …”, “the government considers it to be appropriate that …”
At a minimum all public servants should have the right to refuse to follow directions to
present information in ways that mislead the public in the service of partisan interests.
Furthermore, if they are directed to work in ways which are contrary to legislation or other
published rules on probity, they should be able, in fact obliged, to refuse such work and to
provide a signed statement as to why they are refusing, without any detriment to their own
interests. Such a situation may arise, for example, when ministers seek to make a grant that
does not align with legislation, or that has not been subject to required assessment.
Along with a right to refuse such work, there should be more obligations of responsibility
placed on public servants to account for their own work. There is an established, but absurd
tradition of responsibility-shifting in many public service departments, where people sign off
on their subordinates’ work, even if they have had no part in the task, and , conversely,
where public servants get their bosses to sign off on their own work.
Public servants should not be expected to write speeches for ministers, or to prepare press
releases or correspondence for ministers with partisan spin. Ministers have their political
staff who can add spin to briefs prepared by public servants.
Public servants should always be conscious that their salaries and administrative expenses
are appropriated by parliament, not by their ministers, and that they are “public servants”
rather than “ministerial staff”.
- Leadership – not the same as authority
It is easy to confuse leadership with the exercise of authority, but they are separate
The authors of the review have written about “a leadership model”, but with its focus on
the roles of departmental secretaries and departmental heads, it is really about the exercise
This is not so suggest the authors are wrong. Every organization needs an authority
structure. But leadership is something different, and it does not always mesh neatly with
the exercise of authority. Leadership is not necessarily associated with any position, nor is it
necessarily top-down. Leadership can be exercised from any position.
The authors correctly identify authority vested in departmental secretaries and agency
heads, but there are other sources of authority, including the law and professional
standards as mentioned above. Just as the CEO of an airline does not have the authority to
direct pilots to disobey air traffic rules or to depart from professional judgement on matters
Comments on Review of the APS 9 of safety, so should departmental secretaries and agency heads understand that they are
Leadership, particularly in the public sector, is another matter. To paraphrase the work of
Ronald Heifetz, professor of leadership at Harvard’s Kennedy School of Government,
a set of activities involving the mobilization of the resources of an organization or
people to make progress on the difficult problems it faces7.
Leadership is not confined to any one person (Heifetz avoids use of the term “leader”). It
does not necessarily involve the issuing of orders (that’s the task of those in authority).
Rather, it’s about helping people deal with painful adaptive change, which is why his theory
is particularly relevant in the context of public policy. An example may help illustrate
Tariff reform – a case study in public sector leadership
Economic reform during the Hawke-Keating administration (1983 to 1996) provides a case
study of Heifetz’s theories in action.
By the 1970s most economists agreed that Australia’s policy of high tariff protection was no
longer serving its purpose, and was freezing the country into an uncompetitive and
outdated economic structure. In the 1973 Commonwealth budget the Whitlam Government
introduced a 25 per cent tariff cut, but such was the reaction among those affected that
over the next ten years, under both Labor and Coalition administrations, there was no
further progress on tariff reduction.
The Whitlam tariff cut was a clear exercise of government authority, but in failing to bring
the affected parties on board, and in failing to establish momentum for further reform, it
was a failure of leadership.
The approach of the Hawke-Keating Government was quite different, and in line with the
Heifetz model of leadership. The industry minister, John Button, initiated a process whereby
many people would be called upon to exercise leadership by helping those affected by
possible changes in policy – workers and investors in manufacturing – understand the need
for change. Those included staff in his own department – not only senior staff, but all staff in
policy-related areas who would meet with businesspeople in their own premises, learning,
asking questions, explaining, exploring hypothetical options, rather than giving directions.
Union officials, staff of lobby groups were involved. Independent academics who had no
agenda of self-interest were involved. Often in roundtable discussions those in the most
senior positions, including the minister, would leave it to people down the line to make
suggestions, as those in high authority have the burden of an expectation that they have to
be firm and resolute. As Heifetz points out, positions of authority have certain assets that
allow for the exercise of leadership, but they also have certain liabilities.
Ronald Heifetz, Leadership Without Easy Answers Harvard Belknap 1994.
10 Ian McAuley
The process took many years, and there was pain for many workers and investors. But that
work having been done – the work of leadership – the government was able to exercise its
authority and legislate a program of tariff reductions.
It is informative to compare this approach with the more recent approaches taken on two
important issues – implementation of a resource-rent tax and development of an energy
policy. In spite of a copious body of economic research supporting such policies, both failed
politically. Leadership has been found wanting.
It is possible that the governments involved would have succeeded had they followed a
process more in line with Highways’s model, and have used the public service as a means of
connecting with the community, well before developing any specific legislative proposals.
A strong presence outside Canberra would have helped – a problem identified almost a half-
century ago when the Combs Commission reported. More recently that problem has been
manifest in the failure of the home insulation scheme, which was managed by Canberra-
based senior public servants, with little technical knowledge and no connection with
regional industry networks.
The Combs Commission made a number of recommendations in relation to the regional
reach of the Commonwealth public service. Some of their recommendations are dated: for
example, with lower-cost travel there may be less need for reliance on a permanent
physical presence in the regions, but there is still a need for more contact with the
communities affected by, and who can help shape, public policy.
The need for such presence and consecutiveness has probably increased since 1976,
because since then there has been a significant concentration of lobby groups in Canberra.
While some lobbies can represent broad interests, they often are dominated by established
interests, and some seem to be more concerned with sustaining their own legitimacy than
with representing their members. Public servants whose main contacts outside their own
departments are lobby groups are easily subject to the bias of regulatory capture.
In years of consulting for Commonwealth, state, local, multinational and foreign
governments, I found Commonwealth public servants to be very detached from the people
whose interests they are dealing with. State public servants, while often having less formal
qualification and less sophistication in ideas of public policy, seemed to have much more
awareness of how government policies take effect in and are received by the community.
Also, they tend to have more administrative discretion and delegated authority than their
If the Commonwealth’s role were confined to a minimum set of functions necessary for
federal-state coordination, isolation in the so-called “Canberra bubble” may not be a
significant problem, but ever since the reforms of the 1940s both Coalition and Labor
governments have been becoming more involved in functions that were once state
preserves. If the Commonwealth is to be so involved it should have much more presence in
the states, at a practical working level and not just through high-level bodies such as COAL.
Comments on Review of the APS 11
This is not to make a case for decentralization of administration. For example, moving the
Pesticides and Veterinary Medicines Authority to Armidale solves no problems, because
Armidale is no more connected with the rest of Australian than Canberra is, and, of course,
it makes coordination with other departments more difficult.
Rather, it is about reaching out to the community, in realization of the “public” aspect of the
public service, and in the process contributing to good public policy.