DIRDC secretary Steven Kennedy’s thoughts on the role of the APS

By Steven Kennedy

Sunday September 30, 2018

Dr Steven Kennedy is the secretary of the Department of Infrastructure, Regional Development and Cities.

This is the second half of a speech hosted by the Australian National University College of Business and Economics on September 24. The Mandarin has separately published the first half, in which Dr Kennedy addresses global challenges.

The capacity of the public service to understand the forces shaping the Australian economy and society is essential if we are to effectively advise governments.

However, a high performing public service needs to be able to do much more than consider global trends if it is to effectively deliver Government programs, administer laws and regulations, and run operations. So it is a good time to discuss the future of the public service given the review being conducted by David Thodey, Maile Carnegie, Glyn Davis, Gordon de Brouwer, Belinda Hutchinson and Alison Watkins.

The review represents a wonderful opportunity to analyse, and re-set where necessary, a federal public service that for over 100 years has served Australians well, through 30 Prime Ministers and their governments. It is worth briefly reflecting on how the federal public service has changed over the past 40 years.

Forty years ago, in 1978, there were 159,000 public (federal) servants representing 1.1% of the population and government expenditure was 25% of GDP. Of that expenditure, welfare expenditure represented 28%.

Apart from versions of most of the same departments we have today, there was a public employment agency, Medibank was being disbanded with a continued reliance on private health insurance, there was a GBE called Telecom Australia which had been founded 3 years earlier, the federal government owned and operated hundreds of airports and a range of other Government business including banks.

Defence expenditure represented 9% of government expenditure and was supported by a defence industry strategy. Interest rates were set by the Government and the exchange rate was managed. Australia’s population was 14.4 million.

Twenty years ago, in 1998, there were 121,000 (federal) public servants representing 0.65% of the population and government expenditure was 24% of GDP. Of that expenditure, welfare expenditure represented 36%. Apart from versions of most of the same departments we have today, employment services were outsourced, Medicare had been established as bipartisan policy, the airports had been sold or leased, Telecom privatised, and a range of other Government businesses had been sold.

Defence industries represented 8% of government expenditure and capability was underpinned by an off the shelf approach to purchasing. Interest rates were set by the RBA in response to a policy set by Government and we had a flexible exchange rate. Australia’s population was 18.7 million.

Today in 2018 there are 152,000 public servants representing 0.6% of the population and government expenditure is 25% of GDP. Of that expenditure, welfare expenditure represents 36%. Apart from versions of most of the same departments we have today, Medicare is still bipartisan policy, there are three new GBEs among others, Western Sydney Airport Co, the National Broadband Network and Moorebank Intermodal Terminal. These three GBEs happen to be in my portfolio.

The National Disability Insurance Scheme has been created and all things given will increase the scope of government by 1.3% of GDP by 2044-45. The airports are still sold or leased but there are calls to regulate them more robustly. And we have seen a bipartisan approach to the biggest planned investment in defence capability for many years underpinned by a strategy to re-establish an Australian defence industry. Interest rates are still set by the RBA in response to a policy set by Government and we still have a flexible exchange rate. And today the population is over 25 million.

Compared to 40 years ago, the public service is expected to be more adept at providing services, not only to keep pace with modern expectations of service delivery but also because of the growing number of Australian we are expected to serve.

For example, for income support recipients alone, ignoring many other payments such as family allowances, supplementary payments, family tax benefits etc, the number of recipients has increased, with population, from 2.1 million in 1978 to nearly 5 million in 2018.

Further, the capacity to establish and appropriately govern institutions such as the NDIS and government companies such as the Western Sydney Airport Co, is being called upon more than in the past.

The use of institutions (bodies that sit apart from departments and often established through legislation) to deliver government policy is a long term trend that has served governments well.

My interest in institutions stems from own experience, spending a number of years working at the ABS. From my own experience, I have seen institutions such as the ABS, RBA, PC, ACCC, and in the infrastructure and transport portfolio AMSA, CASA, Airservices Australia, ATSB and Infrastructure Australia, serve governments highly effectively.

These institutions allow us to build sustained capability in a way departments can struggle to do. Much of the research I have drawn upon this evening has been undertaken by institutions or, in the case of my department, from our standalone analytical function, the Bureau of Transport, Infrastructure and Regional Economics.

Institutions are not without fault — their greatest strength is their biggest weakness, from expertise and uniqueness of role can arise inflexibility, arrogance and insularity. But in the main, our institutions manage these tendencies well enough.

The pressures on departments to serve governments across the torrent of issues governments face on a daily basis are substantial. Departments have a much harder time of it than institutions and this has both time and content dimensions. I should say that I don’t regard the increasing load of short turnaround advice as a distraction from the real work of the public service, it is another and real demand of government that the public service has to respond to, and we can do that through capability, culture and structure.

Let me finish by making a couple of quick comments on some of the issues that come up in discussions on the federal public service review.


There is a suggestion in many peoples’ conversations about the public service that there has been a loss of capability. I don’t agree with this. For example, the quality of graduates the Service takes in every year is of the highest order. We continue to have little trouble attracting committed and motivated staff, with excellent academic credentials. This has been the case for many years. From this starting point, the service has built a highly capable workforce.


With such a wide variety of tasks to deliver for governments, the public service has to develop deep expertise. As noted earlier, one way to do this is through public institutions rather than departments, but departments also need this expertise. There are many aspects of this question and too many for me to cover this evening but let me mention one: the acknowledgement that we need to invest in groups of professional services, such as analytical services, to better understand changes in society in order to advise governments effectively.

A colleague of mine once suggested to me that the approach to expertise in the public service must be that of a serial expert. Public servants must spend sustained time on issues, they must be trained, and they must engage with business, the community and academia. This avoids the curse of the gifted amateur where smart and capable people not married with deep knowledge make unnecessary mistakes by not seeing the subtleties.

Risk Aversion

The pressure on public servants to get it right is immense. Today, it seems to me that the expectation is that we get it 100% right 100% of the time, and there are many different views of what ‘100% right’ is!. The response to this heightened expectation can’t be to take less risk or to not take the appropriate level of risk in pursuing policy.

Instead the approach has to be to open up a conversation with the community about why the risk is worth taking, the returns that can be derived, associated with an honest conversation about when it doesn’t come off. Easy to say, harder to do.

Fragmentation or Silos

This is an issue both within and across departments, and between public services. Let me make a couple of remarks about the latter. There is a need for the federal public service to be far better connected to state and local government public services.

For example, to deliver on policies that address issues of a spatial nature, such as cities policy, is crucial. Mechanisms to promote significant movement of public servants between levels of government would be a good start. Many of the policy issues we confront are complex, multifaceted and require the skills of public servants familiar with the challenges of all levels of government as well genuine engagement outside of the public service. Nic Gruen has an insightful take on these issues through his characterisation of policy problems as thick and thin.


The public service’s ability to execute its many functions relies on trust. In most developed countries trust in government is at record lows. This is also the case in Australia as shown in the Edelman index. Interestingly, not all public services or aspects of government are suffering the same decline in trust. For example, OECD research shows that trust in public servants in delivering health services, education and police remains high.

The direct exposure that the community has committed to public servants, such as nurses, is in my mind an important element of this trust.

I have found in my career that the broader group of public servants I have worked with have been just as committed to the public interest as the nurses I trained and worked with more than 30 years ago. There is something special about a commitment to public service. But the community is increasingly sceptical and we have to address this scepticism to be effective.

I think a good start would be more engagement across the board. More engagement with the community, with business and with academics, where we can test ideas, be straightforward about what the government expects of us in our various policy domains, and occasionally take a risk. This is perhaps more important than ever in the face of the global forces I discussed earlier that can disrupt economies and communities.

About the author
Inline Feedbacks
View all comments

The essential resource for effective
public sector professionals