What might become the most important public service reform blueprint in years has been released: Learning from Failure: Why large government policy initiatives have gone so badly wrong in the past and how the chances of success in the future can be improved. For brevity, let's call it the Shergold report.
Subscribe now for unlimited access.
or signup to continue reading
Its author, Professor Peter Shergold, chancellor of the University of Western Sydney among other jobs, is a former secretary of the Department of the Prime Minister and Cabinet and a former head of the Public Service Commission. His review was asked to take account of lessons from the home insulation program and the national broadband network roll-out.
Rather unfairly, Paddy Gourley (see page 6) takes Shergold to task for addressing them; he had to, they were his terms of reference. However, as Shergold notes, these are not the only failures from which lessons can be learned. Problems in implementation of large government programs are common. Indeed, the British government used a sensational title for its documentation of three decades of failings, The Blunders of Our Governments. By contrast, the Shergold report takes a balanced view, noting positives as well as lessons from the negatives.
Although written in a distinctively Shergoldian style, the report is not an idiosyncratic, individual view. It reflects inputs from many senior public servants, credited in an attachment. Moreover, it was commissioned late in 2014, reported in August 2015, but was only published on the Public Service Commission website last month. This timing indicates that the Secretaries Board – the heads of Australian government departments – have been reflecting for some months on how best to deal with the challenges it poses for management and leadership in the public service.
So what does Shergold conclude? Among other things, policy advice should be in writing, not oral, and strictly confidential. Cabinet submissions should include a personal ministerial statement "outlining the policy's purpose, expected outcomes and anticipated implementation risks". Ministerial staff standards should be tightened, and staffers have regular joint forums with public servants to build understanding of their respective roles. There are suggestions on how program management can be improved. The Australian Public Service should be more open to outsiders, with more interchange with the private and not-for-profit sectors. New programs should be trialled rather than rolled out straight to large scale, with seed funding provided for business cases and proof-of-concept stages. The public service should allow contractors more flexibility to take their own approaches to service delivery against agreed outcomes.
The most debated so far is the suggestion that the Freedom of Information Act be amended to increase confidentiality around public service policy advice and opinion. Views are divided. As the report notes, "placing restrictions on freedom of information is extraordinarily sensitive". As a principle, transparency almost always enhances the quality of government: the report notes it empowers citizens, improves service quality, supports innovation, and ensures accountability. Shergold argues, however, that policy advice is different due to politics: "differences of opinion between a minister and a secretary will soon become a political issue if they are made public and can seriously damage the relationship between the two".
Gourley suggests that it's possible FOI makes policy advice better: it allows bold, well-argued advice to shine. There is some evidence for this. A British parliamentary inquiry revealed the advice on which the Blair government chose to join the Gulf War was based on what public servants thought ministers wanted to hear rather than facts, evidence or sound analysis. One of the (many) reasons it was bad advice was because it was hidden behind a veil of secrecy.
On the other hand, in Australia's combative political environment there is huge potential for mischief to be made from exploiting perceived differences between ministers and their public servants. And ultimately, if senior public servants believe the possibility of public release is dangerous – whatever the evidence says either way – they will censor their own advice. Shergold believes they do currently: the report says "agency heads have gone on the public record to say that the potential for public disclosure is constraining advice to ministers".
I have argued in the past that the value of policy advice being made public is that it would improve the woeful standard of public policy debate in this country. This report has changed my mind. However, if advice is to be kept secret, there needs to be other ways to expose ideas and ensure quality. If incoming-government briefs – some of the more important pieces of policy advice – were no longer accessible through FOI, an alternative worth considering would be for each department to publish, post-election, a summary of the important questions it faces (social, economic, demographic, environmental) and summarise possible options for the future, without indicating what direction the department had suggested its new minister take. Such documents would complement the pre-election fiscal and economic outlook, enhance public understanding, and would require little extra effort given that the base work would already have been done for the incoming-government brief. Other mechanisms include traditional Westminster system "green" papers (policy discussion and information papers to help improve debate – it would have been helpful to have had one in recent tax debates) or more innovative mechanisms such as the think pieces or "provocation" papers commissioned by the Victorian government.
An approach to avoid the problem of departments hiding poor advice behind secrecy is, as the report suggests, greater accountability on secretaries for the quality of their department's policy work. It suggests secretaries' annual performance discussions are one existing opportunity; while helpful to an extent, the limitations of time in these discussions means the quality of a department's policy advice is unlikely to examined deeply. A more rigorous approach would be peer review of policy advice by external experts; not costly if done on a random sample basis, and with the potential to identify and correct systemic weaknesses. It is a good approach – the Shergold report itself was subject to peer review.
Other Shergold recommendations are excellent – especially, opening up the public service and encouraging better standards, interaction and understanding among ministerial staffers. Traditionalists will hate suggested initiatives on greater interaction with business, adaptability, contestability, digital-enabled engagement, but they are inevitable and must be embraced.
Finally, the report has a substantial section on creating a risk culture. Shergold is spot on in his diagnosis of the problems created by the current culture of excessive risk aversion. What he does not investigate is whether this culture has evolved as a rational response to the pressures placed on the public service. If it is, more must be done to address the underlying incentives that have led to the present culture. Senior public servants have been talking about the need for more risk taking for many years now, but a new culture is not taking hold. If the Shergold recommendations are to carry the day, actions as well as rhetoric need to change. How to achieve that is worth exploring in depth – so will be covered in next month's column.
Stephen Bartos is a former senior public servant. stephenbartos@fastmail.fm
