‘There but for the grace of God…’ – Sue Robertson’s reflections on leading the APS Integrity Taskforce

The APS Integrity Taskforce was established in February 2023 to provide a ‘bird’s-eye’ view of the APS integrity landscape, identify gaps and look for opportunities to learn from and build upon the important work already underway across the service.

The Taskforce was able to take a balcony view of integrity activity across the APS. It examined culture, systems and accountability, and how the relationships between each one shape human behaviour. It published its final report Louder than words: an APS integrity action plan in November 2023.

Sue Robertson, a public servant, international law expert and former adviser to the United Nations, was appointed leader of the taskforce. In this article she shares her insights into the process of defining integrity, developing key principles to select the issues they would work on, what Robodebt can and can’t teach us, and the limits of black-letter law to addressing integrity issues.

When I began this work, I tended to think of integrity as an innate quality – you had it or you didn’t. Like Sir Thomas More. Or Nelson Mandela. But there are serious limitations in framing integrity as the stuff of saints. I now think of integrity as a skill and a habit. A kind of behavioural dance that we can move in and out of all day, depending on how self-aware we are in the moment.

Good habits guru, James Clear, reckons that we do not rise to the level of our goals, we fall the level of our systems. My team wanted to explore how we could bring this system thinking to the Australian Public Service. At the same time, I did not want compliance frameworks to be the focus of our work. Whenever a failure of public administration occurs in Australia there is a tendency to grab the nearest lawyer and regulate our way out.

But more law cannot solve ethical failures. The Australian public service has a surfeit of laws, codes of conduct and Values (including the much-undervalued APS Commissioner’s Directions on the meaning of the Value of Ethics) and whilst the ANAO has rightly indicated we need to improve here – the more interesting question for our Taskforce was what gets in the way of people doing the right thing? How could the practice of leaders encourage the behavioural integrity we wanted to see both for individuals and the public service as a whole?

The Policy Orbit

The APS Integrity Taskforce was established by the Secretaries Board against the backdrop of numerous failures in public administration and in a crowded policy space. Integrity was Pillar 1 of the government’s APS Reform agenda and the National Anti-Corruption Commission had opened its doors in July 2023. We were standing on the shoulders of extensive ongoing work by committed people and tomes of historical review recommendations. The Thodey Independent Review into the APS. The Bell Inquiry into the former Prime Minister’s appointment to multiple ministries. The Pink Batts Learning from Failure report- the list was long.

We had a broad Terms of Reference. In essence our task was to take the ‘balcony view’ of integrity activity across the APS, and ask what else? Why? What does this ‘busyness’ on integrity achieve as a system? We were tasked to report to Secretaries Board with a comprehensive plan of action in six months and an interim report in three months.

The key stakeholder interests in integrity were reflected in the expertise and backgrounds of our small Taskforce membership – two people from the Attorney-General’s Department, two from the APSC, one from the Australian Law Enforcement Integrity Commission (who helpfully had been responsible for the Commonwealth Integrity Maturity Framework) and one person from the Department of the Prime Minister and Cabinet. This diversity of experience allowed essential bridges of trust between our mandate and that of our APS stakeholders. I kept the Taskforce size small to keep us focused and prevent overreach on the breadth of the final report.

Building relationships of trust was crucial since our work placed us squarely in the territory of other department’s responsibilities. We needed to avoid provoking defensiveness and being viewed as an unhelpful distraction in stretched policy times. What helped here was for the Taskforce to adopt an attitude of ‘walking beside’ agencies to support and amplify their integrity work. We made it clear we were not out to shame any department, rather, we could highlight their good practice and hopefully add some new ideas. The fact that the Taskforce was working directly to the Secretaries Board also gave us a convening power with our partners because ultimately all departments had a stake in the final report recommendations and wanted their leaders to be prepared and able to talk to the issues.

The story we choose to tell – the lessons we choose to learn

The Robodebt fall out was not the Taskforce’s sole reason for being. Adopting the phrase of a former Australian Prime Minister, it doesn’t explain everything, it doesn’t explain nothing, it explains some things. Our terms of reference covered the APS as a whole, but Robodebt provided important context to our thinking. Since our Taskforce wound up, the Royal Commission issued its report and the government has agreed to all of its 56 recommendations in whole or in principle.

During the life of our Taskforce, the public service witness testimony coming from the Robodebt Royal Commission was a slow-moving train wreck. But what integrity lessons would we glean? Was it a story about political delivery at any cost? Was it a story about law not being respected in the policy process? Was it a story about the human impact of bad policy? Was it a story of the lost art of candid advice to government? This mattered. How we interpreted the story would shape the solutions we put forward.

One thing was clear to me though, the pressures revealed by Robodebt testimony were not foreign to me as an SES leader. The thought, ‘There but for the grace of God go I’ came to mind more than once. The Robdodebt debacle had lessons beyond the departments involved. Channelling Swiss psychoanalyst Carl Jung, perhaps Robdodebt revealed to the APS its own ‘shadow’ self – what happens when we overplay our strengths of responsiveness and relinquish our role of stewarding the long-term public interest.

At another level though, Robodebt was not a great example of the complexity of ethical challenges our leaders face. It was too blatant, too obvious. Following the law is meant to be ‘the floor not the ceiling’, of our conduct. The bulk of ethical dilemmas faced by leaders are in the grey zone of ambiguity with more than one reasonable answer. So, there were risks in making Robodebt the example from which to base an integrity discussion.

Three months into our task, the PWC scandal broke revealing the corporate misuse of confidential tax information from Treasury. This magnified existing concerns about the dominance of consultants in the public policy process diluting APS skills, and the potential gap in values between the private and public sectors -profit and growth, versus the public interest. Secretaries Board were also keen to better manage the risks of conflict of interest arising from employment mobility between the public and private sectors. Multiple Senate committee inquiries and independent reviews were established while the Central departments kicked into overdrive to ‘fix’ the problems and reassure the government and the public they were responding.

In short, the APS Integrity Taskforce faced high expectations to provide new ideas for tackling age-old integrity challenges of very different kinds, agency reactivity, and moving media plates under our feet which required us to pivot on more than one occasion.
On one level, the political momentum to action on integrity was a force to harness. Our questions and outreach were acting as a catalyst for agencies to get ahead of the integrity recommendations that were coming down the track. On another level though, my ‘delivery’ brain hankered for a cohesion that did not exist and wondered if there would be anything left for the Taskforce to recommend after the dust had settled on these scandals.

Would our report contain anything original, or more importantly, useful? Like other work done in the Department of Prime Minister & Cabinet which holds a shadowing policy function, the Taskforce was responsible for recommendations which we did not have the policy authority to implement. The quality of collaboration determined the success of the work. We could put forward our best advice to Secretaries Board, but without getting those responsible for implementation on board, our work would be of little utility. We would revisit the metrics for success of the project more than once.

How long is a piece of integrity string?

There were a series of questions in play for the Taskforce that had to be tackled early:

• How much to focus on the individual versus systems aspects of integrity?
• What was the best mix of formal and informal approaches to building integrity?
• How specific should we get on definitional debates?

On definitions, Integrity is a deep well. Over the ages, it has been pondered by Stoics, codified by lawyers, and more recently, enforced by police and auditors. For modern public administration, it is the mortar in the bricks of public trust in government. In the APS we have our legislated Values and the excellent (though not well-known) APS Commissioner’s Directions which unpack the Value of ethical behaviour, including integrity.

Unsurprisingly, as a matter of practice, modern agencies frame integrity differently in accordance with their function. A Cyber security expert will say integrity is building systems impenetrable from hostile interference. Statisticians see integrity as data accuracy. Police have an integrity lens on the pointy end of corruption and the criminal law they are enforcing. Procurement professionals focus on probity, transparent process and preventing conflicts of interest in the tender process. Human Resources professionals bring both case management and values perspective to organisational integrity.

This meant that any discussion with stakeholders on integrity had to avoid assumptions and start with explicit clarity on what we were talking about. At the same time, I knew my legal training could make definitional debates endlessly fascinating to me and a potential sinkhole. However, the team only had six months to complete this task and getting lost in definitions could suck up too much time for little reward.

What is in scope also necessitates important decisions about what work is left out. There were two other issues keeping me awake at night. They were important but I knew it was not feasible to address them in the six-month time frame. Any policy change worth its salt needed to be based on evidence. The APS needed to know the baseline state of our integrity landscape and, departments needed to know if they were improving over time. So, data was on my mind. It had also been a big feature of previous work done by the APS Reform Office. What data to collect, how to collect it, and how to interpret the different ‘stories ‘that data can tell both across the service and within departments. In the result we made one recommendation on data which acknowledged the issue and further work to be done.

The other fact was the pace at which Language Based Machine Learning Artificial Intelligence was being publicly released. Action to improve integrity was well and good in the APS, but what integrity and public trust mean in an environment where information could not be verified as true? Both would be consciously left for other greater minds to continue their work on.

The team and I develop a series of principles to help us select the integrity issues we would focus on and the assumptions that would shape our work. These were that:

• Integrity at its heart is ethical behaviour, trustworthiness, following through on our agreements and acting lawfully with public resources.
• But integrity is more than following the law, doing the right thing requires ethical leadership and a longer-term focus on the concept of the ‘public good’. (The Commissioner’s Directions frames this as ‘acting in a way that is right and proper as well as technically and legally correct’).
• Integrity has entwined individual and organisational elements We needed to address both. People with courage and good intentions can be crushed by systems that make the cost of doing the right thing too high.
• Accountability and compliance systems are necessary but not sufficient to promote ethical behaviour. They are also too late. Our focus would be on the fallible human element – how do we uplift ethical leadership and better understand the psychology of behavioural ethics and psychological safety?
• Whilst integrity is everyone’s business in the workplace, our recommendations would focus on the APS leadership. Without leaders walking the talk, integrity talk would just breed more cynicism.
• Most people who choose to work in the public service come to work wanting to do the right thing. (The small cohort of people who intentionally want to break the law were not our focus.) We wanted to focus on the interplay between culture, systems and human behaviour to help people do the right thing.
• We would not reinvent the wheel on integrity, we needed to build on what had gone before. Some recommendations would likely amplify existing work and seek to strengthen it.

Building the plane as you fly

Our teamwork method began with developing a series of rapid-fire research papers on the key integrity topics we had selected. The team developed its own template of questions the research papers would seek to answer. This allowed us to quickly get across the issue by defining the problem statement, understanding the state of play in terms of reform or policy activity, identifying the issues’ relevant legal or regulatory frameworks, and brainstorming how to amplify or strengthen the reforms. We also had a section on cross-links to other major reviews and identified the ‘known unknowns’ for follow up. Taking the lead of the OECD’s Strategy for Public Integrity we decided to structure our recommendations into three areas of Culture, Systems and Accountability.

While the team prepared their research papers, I did a high level, literature review of the key Australian reviews and books of interest. My focus was obstacles to ethical behaviour rather than regulatory reform or anti-corruption. In our broad approach, I was particularly influenced by Professor Marianne Jennings, The Seven Signs of Ethical Collapse (2006). Although her study focused on the US private sector, the lessons she outlined of how to spot moral meltdowns had clear analogies for the public service.

Her analysis identified the risk of taking an overly legalistic approach to ethical failures and the importance of moving staff thinking from ‘Could we, do it? to ‘Should we do it?’. Of her ‘Seven signs’ predicting collapse, four peaked my interest for the APS: the ‘pressure to maintain those numbers’ (this focused on private sector growth but I could see parallels in Robodebt with political pressure to maintain the budget benefits of the policy); the risk of ‘fear and silence’; managing ‘conflicts of interest’; and ‘goodness in some areas atoning for badness in others’ This final ‘Sign’ recognises the human bias to justify dodgy activities because of the overall noble cause being engaged in (she identifies NGOs as particularly vulnerable on this count.) I wondered if the altruistic aims of the APS tempt us down the path of the ends justifies the means.

Professor Jennings’ research is a clear precursor to Amy Edmondson’s later work on psychological safety and how it contributes to ethical behaviour and the ability to report failure (Edmondson, Ferrére, Rider, Renerte, 2022). Psychological safety is the belief that an environment is safe for interpersonal risk taking. Edmondson’s data demonstrates that cultures where people feel able to contribute ideas, share information and report mistakes are more likely to engage in ethical behaviour in the workplace. Psychological safety has other benefits as well in terms of creativity and inclusion of all team members, so it is a no-brainer for the public service to build this skill.

Interestingly, in our outreach with leaders on psychological safety, a couple of reoccurring themes surfaced. One was a mistaken belief that psychological safety was about being ‘nice’ which would make tough conversations in the workplace more difficult if we mollycoddled people. In fact, the opposite is the case. The presence of group trust and psychological safety is necessary in order to have the uncomfortable conversations we need to have in the APS. Another implicit thread that emerged was whether an emphasis on speaking out would open the floodgates for peoples’ individual ethical dilemmas, unhelpful dissenting views and create an expectation that all of these would be acted upon. Would this get in the way of just getting stuff done and our role as impartial public servants?

These responses revealed the pressure for delivery our leaders are currently under. I too wondered if leaders, myself included, had the inter-personal skills and time to navigate this uncomfortable terrain with sufficient dexterity. But I was convinced it was worth it. Yes, emphasising psychological safety could raise more issues from staff, but as Psychologist Brené Brown has astutely observed: “Leaders must either invest a reasonable amount of time attending to fears and feelings, or squander an unreasonable amount of time trying to manage ineffective and unproductive behaviour”. This has proved true in my experience again and again.

The integrity conversation required us to be more explicit with our teams about framing the purpose of our work and acknowledging the uncertainties and tensions that exist in achieving the work. Two other tools are essential to building psychological safety, inviting genuine participation from our teams, and responding productively to the risks that staff take. We have all had experience putting ourselves ‘out there’ in a group discussion to find only disinterest and crossed arms, or worse, having our heads chopped off. Lesson? I won’t be doing that again. No wonder people do not want to report mistakes early.

The team’s integrity themed research papers were used to socialise ideas with our key stakeholders over six months. We predominantly focused on APS departments and agencies due to the timeframe. Several departments were already doing great work on integrity culture and my team looked to hoover-up their good practice to see if it could be shared across the Service. Of the external discussions, talking with The Ethics Centre and the Cranlana Centre for Ethical Leadership proved insightful. Both highlighted the role of critical reflection and pausing needed to make good decisions. This also linked in my mind to Daniel Kanheman’s work on modes of thinking in Thinking Fast and Thinking Slow (2011). A reoccurring question was how we could craft the practice of THE PAUSE into our daily leadership when we were moving apace? This would allow us to manage our ancient, learned stress responses which prevent us from thinking with ethical clarity.

Legal advice – the floor not the ceiling

The Taskforce needed to address the importance of legal advice and record keeping in our Systems recommendations. Robodebt, amongst other things, had demonstrated what could happen when inconvenient law was sidelined by policy leaders. Sadly, there needed to be a back-to-basics reminder about legal scaffolding for all Commonwealth action. (Along with the obvious question, does this policy actually work?).

At the same time, the Taskforce was cognisant of unintended consequences of our recommendations. We didn’t want our advice to send the message that public servants should not engage with legal risk – an inherent part of the policy process and a political reality. We want, and need, our APS to exercise its judgement and critical thinking in relation to policy risk. When Commonwealth action crosses a legal bright line (as it appeared to in Robodebt) that advice needs to be placed squarely before Ministers and better solutions provided.

More often though, public servants face a spectrum of legal risk depending on the kind of activity under consideration. The Taskforce held some important discussions in the APS on the integrity issues that arise in framing legal advice purely from ‘risk’ lens, particularly when advising on the lawfulness of Executive Action. For example, an overly flexible approach to legal advice in terms of ‘risk’ rather than on the question of lawfulness can dilute the clarity of legal advice to Ministers, and inadvertently absolve the APS of its rule of law stewardship. As APS lawyers, it is important we give our best legal view, and this is a separate question to the risk of being taken to court. In the result, our Taskforce advice strove to balance these concerns and recommended that accountable authorities provide a direct line of access to the Heads of Legal to ensure their advice figured appropriately in decision making.

Pens down

The APS has a tortured relationship to record keeping. Despite the existence of obligations, extensive guidance, and training, it turns up in APS reviews as the ‘problem child’ again and again. Two things are true about writing things down in the APS. Firstly, it is necessary for accountability of key decision points and transparency of corporate memory in policy making. Secondly, it can embarrass the political class. Public servants serve the community through their ministers and without the trust of their Minister’s Office, their influence is minimal. Whilst public scrutiny is crucial to the democratic process, it has consequences for how public servants approach their work.

The Freedom of Information (FOI) law rightly favours transparency in the public interest and explicitly rejects political embarrassment as an exception to disclosure. At the same time, the ‘gotcha’ 24-hour media cycle is hungry for slip-ups. We also seem far removed from the original intention of the FOI framework when Senate Estimates briefs and every piece of paper public servants carry is being subpoenaed in the name of transparency. Improvement here is not about reverting to copious, verbatim note taking of every step we take as public servants. Rather it is about a common-sense judgement of what needs to be recorded in line with our public service role. This role needs to be better understood with ministers and their offices.

Any recommendations on reform of FOI are fraught. I agree with Peter Shergold that the APS and the political class need a confidential space to wrestle down development of good policy, including the ability to change their mind and not be accused of ‘flip-flopping’. Unfortunately, this is tough sell politically. Politicians would need to stand up and confidently make the case for broadening exemptions from FOI whilst risking inevitable cries of hypocrisy from the public. Our Taskforce has recommended that the FOI law be reviewed following the current Senate Inquiry to explore if it is contemporary, fit for purpose and meets its original intention. However, larger scale reform seems unlikely any time soon. The government has made clear in its press conference response to Robodebt that there is currently no appetite to amend the FOI Act (at least in response to the Commission’s observation that the exemptions for Cabinet in Confidence should be re-examined).

Working with Ministers and their offices

Although our mandate was focused on the APS, no work on integrity across the APS could take root without tackling the heated reality of the Minister’s office. Our recommendations were being developed at the same time as the significant institutional changes from the Set the Standard Report including the establishment of the Parliamentary Workplace Support Service. Anyone who has withstood the white-water up on the Hill knows the pressure to deliver, the tempo of decision making and the imperative to build trust with their Minister. It can be all consuming and easy to lose perspective. As we have seen from Kate Jenkins’ report, the way in which power is exercised and the power imbalances that exist can make Parliament House a perfect storm for integrity and ethical meltdown.

The APSC Strengthening Partnerships program was an important start for training APS in working with ministers, but there were limitations if ministers and their staff were not part of this conversation. My view is that more needs to be done by the senior leaders of the APS to bring the PMO and other ministers on board this discussion. Mutual understanding on role clarity is key. Most ministers appreciate frank and fearless advice and want to avoid surprises and making policy mistakes. However, new advisers and their ministers need to understand the role of the public service and public servants need the inter-personal skills to communicate risks and solutions to their political masters in a useful way.

The Taskforce highlighted the importance of supporting APS staff working at Parliament House to know the boundaries of their role and to get advice when they were unclear. When conflicts arise in the minister’s office, SES need to have the self-awareness to pause, understand the situation and ‘back’ our APS staff when tricky issues arise. Stewardship as a new APS Value also has an important role to play. Alongside giving ministers impartial advice, stewardship is the invitation to play the longer systems game, to understand the consequences of a decision or action for the future and to ensure we understand who and what else will be affected by it.

The power of informal conversations

Out and about talking to public servants one thing became clear. People were keen to talk about integrity. They were saddened and concerned by the testimony that came from the Robodebt Royal Commission. This told us that there was more room for candid, confidential sharing between senior leaders of the complex integrity and ethical dilemmas they were facing and, a need for more peer support, shared experience, and lesson learning. Training as the regular salve for public service failures was not the answer. Rather, we needed to think about learning in a new way, in an applied way reflecting our real-world pressures. This would require leaders to express greater vulnerability and share their willingness to learn from mistakes.

Compliance approaches to integrity work when issues are clear cut and legal bright lines obvious. Most decisions faced by senior public servants though are in the grey where reasonable minds can differ. We need frameworks to respond to both. We know from the work of Professor Mary Gentile in ‘Giving Voice to Values’, that giving leaders the opportunity to mentally rehearse ethical responses to predictable work challenges before they occur, can increase the chances of better decision making when values are tested. The APS Academy Masterclass Model had begun important work in this direction, and we strongly supported this model in our recommendations.

Appointing the right people and rewarding the behaviour we want to see

Human beings respond to incentives. The Taskforce wanted to underline the crucial signal we send in who we appoint and promote as leaders in the APS. No matter how much ink is spilt on values, or glossy brochures on integrity we know that staff will take their behavioural queues from the leader. Leaders who intimidate and bully their staff cannot build pro-integrity cultures in the workplace. Whilst leaders need to hold people accountable when bright lines are crossed, fear as a rule shuts people down and blocks integrity red-flags from being raised. So, a key recommendation was to appoint leaders who get things done in a way that does not leave a trail of human debris, stress leave and trauma.

The Taskforce worked closely with the APSC on its new SES Performance Framework. This is a crucially important policy change to leadership expectations across the Service. Why? Because it shifts performance appraisal from pure delivery of results to behaviours and how a leader enables their best asset – their people. The Taskforce also lobbied for leaders in performance discussions to be required to demonstrate how they are building psychological safety in their teams. The proof in this pudding would be in the implementation. Can we hold more meaningful performance conversations, tackle uncomfortable issues, and engage in self-reflection without becoming defensive?

Coming full circle – So What?

I said at the beginning that integrity was not the stuff of saints. For me, this helpfully knocks it off the moral pedestal into the imperfection of every day decision making. Integrity is a concept that presses peoples’ buttons and can easily invoke fear responses. My hope is that our teams, our divisions, and our departments can normalise discussions about ethics in our decision making. Leaders can start by explicitly invoking the values which underpin their decisions and making the link to our role in serving the public interest. The biggest game changer in my learning has been understanding how our ancient brain works. I’m noticing when my stress responses tip into defensiveness and close down discussion with my team instead of opening it up. Teaching the APS the neuroscience of building psychological safety in their teams will be central to a stronger integrity culture.

Robodebt showed that the APS needs to remain vigilant on the basics of our impartial advising role and defending the wise constraints of law. But most of our leadership choices are 49-51 either way. The difference for a good (or a ‘least bad’) decision is the quality of our pause before leaping to action. This is something we will need to practice in the rapids of our policy making environment.

The Taskforce worked with many APS leaders committed to doing the right thing for the Australian community. Integrity work is never “done”, it is a skill to be learned and a habit to be rewarded. I’ve been proud to lead the Taskforce’s contribution to this ongoing journey. Secretaries have agreed to return in six months to report on implementing our recommendations. I will be very curious to hear the progress.