Not that everyone is always referring to the same aspect of government practice. Sometimes, it's a feature of cabinet government; at others, a version of ministerial responsibility or an aspect of a career public service. People might not always agree on exactly what Westminster is but they agree that it's in decline and being replaced by something worse.

This suggests two features of "Westminster" as commonly understood. First, it tends to be invoked as a marker of cherished values or practices that are perceived to be under threat. Second, opinions vary about what these values and practices actually are. Can we give any actual content to the concept, other than its function as an indication of traditionalist outrage? Drawing on the work of scholars Rod Rhodes, John Wanna and Pat Weller, we can understand Westminster as a set of interrelated traditions, which are grounded in 19th-century British practice but constantly revisited and revised to suit present-day preoccupations.

From this perspective, the pursuit of a single, definitive Westminster model is futile. So too is lengthy argument over what should or shouldn't be counted as a genuinely Westminster practice. By way of illustration, half a century ago, Westminster principles were used to oppose two innovations seen as dangerously foreign: the Scandinavian institution of ombudsman and the United States practice of interrogating government officials at legislative committee hearings. At the time, everyone accepted that Westminster conventions of ministerial responsibility to Parliament did not allow for independent, public scrutiny of public servants. The reformers, however, carried the day and ministerial responsibility evolved to accommodate these innovations. Today, no one would seriously suggest that either the ombudsman or parliamentary committee hearings are un-Westminster institutions. At the time, however, the existing understanding of Westminster served a useful discursive purpose. It helped clarify debate by allowing the proposed reforms to be assessed within a valued administrative tradition.

Do any current trends or innovations run counter to current understandings of Westminster principles and values? If so, should we resist them in the name of these principles and values? Or should we accept them as useful additions to administrative practice that don't compromise what we see as the continuing core of our Westminster-based tradition?

I'll focus here on Westminster traditions about the public service, leaving aside other aspects of responsible cabinet government, such as the relationship between prime ministers and their ministerial colleagues. The core values of a Westminster-based public service will be taken to centre around a career-based professional service, appointed on merit and offering impartial service to ministers, including free and frank advice.

Can public servants provide the robust advice ministers need to hear? The near monopoly on advice once enjoyed by Westminster public servants is long gone. Ministers now have many more sources to call on, including their expanded ministerial offices and the growing ranks of think tanks and consultants. Are public servants being crowded out of the policymaking loop? Or can they still apply their unique policy skills and experience effectively alongside other claimants on ministers' attention?

The growing role of ministerial advisers attracts fierce debate. Some see them as hostile gatekeepers filtering out uncomfortable but sound advice. Others take a more benign view of them as complementary partners, serving the minister's political needs while public servants provide the objective argument and evidence. Respectful coexistence is certainly common and is a feature of most well-run portfolios. But it depends on a number of factors, beginning with a solid relationship between minister and department head. Advisers need to be kept in their place and public servants need to be confident that frank advice will find its way on to the minister's desk. In general principle, however, advisers are not incompatible with our Westminster core.

Beside the rise of advisers, other forces may work against the effectiveness of free and frank advice. One is a reluctance among public servants to commit themselves in writing when dealing with politically sensitive matters. This was a lesson drawn from the ill-fated home insulation scheme, which is now the most-analysed program in the federal government's recent history. Two experienced former departmental secretaries, Allan Hawke and Peter Shergold, were among those who examined the reasons for the program's failure; Hawke in the immediate aftermath and Shergold after all the dust had settled. Both agreed that departmental officials failed to keep proper written records and failed to accurately convey serious concerns to their superiors. Shergold saw this as symptomatic of a more general, service-wide deficiency, which other Australian Public Service leaders have since confirmed.

The reasons for this trend away from frank and fearless advice and record-taking are unclear. One alleged culprit is freedom of information legislation, which supposedly discourages frank expression in writing. Several senior public servants recently called for these laws to be revised to allow more room for confidential communication between public servants and ministers. However, the problem may be more in public servants' timid practice than in the legislation itself, which, arguably, allows adequate exemptions. Whatever the reasons, if public servants feel discouraged from being forthright with their superiors, the underlying purpose of a professional public service is frustrated.

Another factor working against the effectiveness of policy advice is the squeeze on departmental resources and the running-down of policy divisions. Continual cuts (so-called "efficiency dividends") have led to hard choices and a natural tendency to sacrifice long-term policy work to preserve more immediate services to ministers and the public. This is short-sighted. The provision of robust, firmly grounded advice is also an essential service.

The practice of outsourcing policy analysis to consultants is no substitute for in-house analysis. Consultants are intellectual mercenaries with little commitment to good policy beyond meeting the terms of the present contract and securing the next one. Public servants tend to have an exaggerated view of the competence of private sector consultants compared with their own. In the home insulation scheme, the Environment Department, out of its depth in implementing a new program at reckless speed, called on several of Australia's leading consulting firms for help. At each stage of the process, the relevant consultants gave the appearance of having everything under control. In the end, however, as the scheme unravelled, they took the money and ran, leaving public servants to pick up the pieces.

This is not to deny a useful role for consultants in plugging occasional gaps in in-house expertise. But extensive, ongoing reliance on outside advisers hollows out the skills base in government agencies, leaving them incapable of exercising the policy oversight that is expected from a professional career public service. In my view, the rise and rise of consultants is the single greatest threat to our core Westminster values. Unless senior public servants and their ministers give greater priority to bolstering the public service's in-house policy capacity, the system will slowly wither on the vine.

The second challenge to Westminster understandings concerns the division of responsibilities between ministers and department heads or secretaries. The traditional Westminster bargain between ministers and senior public servants gave ministers the sole duty to answer publicly for their departments, leaving public servants silent and anonymous. This anonymity decreased when public servants began appearing in public before committees. Westminster niceties, however, were preserved through a division of responsibilities between ministers, who were responsible for "policy", and public servants, who were responsible for "administration". The lines of demarcation were deliberately vague, and ministers could still be called on to take responsibility for administrative matters, such as answering in Parliament on behalf of the department and undertaking to impose remedies.

These conventions are still extremely potent. The later inquiry into the home insulation program focused on the performance of public servants. At the time, however, it was all about ministers. In Parliament, the opposition pursued mercilessly environment minister Peter Garrett and later prime minister Kevin Rudd, even though the main problems were of administrative implementation. Particularly in times of crisis, the conventions of ministerial responsibility remain strong.

On other occasions, however, the newly legislated responsibility of secretaries for their departments' administration has allowed ministers to escape proper responsibility. For example, the infamous efficiency dividends allow ministers to make severe cuts to the public service without taking responsibility for any particular cuts, which are delegated to agency heads. As a result, the formal chain of Westminster responsibility, from departments to ministers, to Parliament and the public, is broken and accountability is diminished. Secretaries need to resist the lure of accepting full responsibility for departmental decisions.

A related area of tension is the extent to which department heads should speak publicly on policy matters. The normal convention – that public servants avoid public comment on policy – remains entrenched, and for good reason. Any perceived support for one side of politics immediately undermines the perception of political neutrality that is essential for a professional public service. Nonetheless, this principle is fraying at the edges. Successive Treasury secretaries, from Ken Henry on, have made occasional forays into debate on economic policy – usually, though not always, criticising both sides of politics. Treasuries are independent sources of economic advice and some slack can be cut to their secretaries in expressing the departmental view. But when Martin Parkinson was at the Treasury, he pushed the envelope further and has continued to do so at the Department of the Prime Minister and Cabinet. Meanwhile, respected former secretaries Terry Moran and Andrew Podger, bemoaning the intellectual poverty of public policy debate, have urged public servants generally to speak out more on policy issues where they have expertise. Ultimately, however, the limits of acceptable speech are set by politicians, who can have long memories.

Which brings us to the third general area: the role of ministers in appointing department and agency heads. All Westminster-based jurisdictions have allowed a greater role for prime ministers or premiers in such appointments, in contrast to traditional views that appointments should be free from political intervention. In Australia, we now accept as uncontroversial that secretaries are on limited-term contracts and the prime minister should have a final say in appointing them. Some traditionalists see these changes as serious impediments to the core values of impartiality and independence, which have increased the timidity of public service advice. Others argue they reinforce a proper responsiveness to ministerial direction and compromise neither impartiality nor the capacity to give robust advice to the government of the day. On this more optimistic view, to which I lean myself, the Westminster core is safe.

More problematic, however, is the question of whether, on a change of government, incumbent secretaries should continue in post or be replaced. The seminal event was John Howard's night of the long knives in 1996, when he dismissed six of the 18 incumbent secretaries. The decision sent shock waves through the public service that still reverberate in Canberra.

One consequence was that Labor, in opposition to Howard, committed to continuing the appointment of incumbents. The party of Gough Whitlam, Bob Hawke and Paul Keating, which had steadily reduced the independence of the former permanent heads, has become the party of Westminster traditionalism. Kevin Rudd, himself a former bureaucrat, ostentatiously continued the tenure of all secretaries, without question.

The issue thus became a matter of partisan dispute. Tony Abbott, as prime minister, felt obliged to respect the precedent set by Howard, his mentor, rather than side with Labor. In addition, many in the Coalition subscribe to the anti-government ideology of think tanks such as the Institute of Public Affairs, for whom the public service is an agent of the progressive left. On this view, the task of Coalition governments is to attack the bureaucracy by reducing its numbers and imposing managers with private sector backgrounds and sympathies. Abbott's response was limited. His dismissals were relatively few – "the night of short knives" as it was called – and his new appointees from the private sector had all had earlier careers in the APS. But he did enough to arouse the progressive left, not just in Labor but also in the media. Progressives embraced the public service as a sacred national institution, along with the ABC and the Australia Council for the Arts. Malcolm Turnbull's early decision to bring back Parkinson, one of Abbott's martyrs, as head of PM&C boosted Turnbull's initial appeal to such people.

The dismissal of incumbent secretaries upon a change of government is an important issue because it undercuts a core Westminster principle: that a professional public service can serve alternative party governments with equal loyalty and effectiveness. Throughout their careers, public servants learn they must suppress their political opinions and fully accept the political direction set by the government of the day. It is a hard lesson, particularly for younger staff who joined the public service with high ideals. But by the time they reach the upper levels, their professional loyalty is usually unquestioned. Replacing incumbent secretaries for the purpose of imposing new policy priorities is a direct denial of this loyalty and deeply demoralising to all public servants – a point recently reinforced by Peter Varghese in his valedictory address.

We should not exaggerate the extent of this threat to Westminster values. Most secretaries survive a change of government to serve out the full term of their contracts. Most new appointees are themselves career public servants who have served both sides of politics. Most successful ministers quickly come to value their departmental officials' support. We are nowhere near the fully politicised system of the US, or the open partisanship of the French and German bureaucracies.

Nonetheless, there are influential pockets of support, particularly in business circles, for the mistaken notion that new ministers should be able to appoint chief executives who are sympathetic to their policy views. In the private sector, after all, such appointments would be standard practice.

More broadly, the current federal political climate, in which the Coalition has been at war with the APS, and Labor has equally vigorously defended the APS, is worrying. However we reinterpret the Westminster tradition, one of its foundations must be that the career public service is equally trusted by all sides of politics. Fortunately, there are signs in Canberra of a possible rapprochement between Coalition ministers and the APS. In NSW, Mike Baird's government has shown that respect for Westminster principles sits well with conservatives. To balance the ledger, perhaps Labor and its cheer squad could be a bit more critical of the APS. It's not that perfect.

Richard Mulgan is an emeritus professor at the ANU's Crawford School of Public Policy. This article is an edited version of his paper, Goodbye Westminster? An Australian perspective, which is scheduled to be presented at the Australia and New Zealand School of Government's annual conference in Sydney this week. richard.mulgan@anu.edu.au