Could it happen to us? Government approaches to learning from the Robodebt crisis
Last week, Michael Brown wrote about how to embed “lessons learned” into policy – in his case on the role of police in the mental health system. Here, I’d like to reflect more broadly on the challenges for organisations learning from failures, particularly where that failure was “someone else’s fault,” or happened “over there,” drawing on the Australian case of Robodebt.
The literature on policy diffusion is replete with examples, theories and frameworks about how ‘good’ policy travels. Many industries (such as healthcare and air transport) are built around learning from others’ mistakes and near-misses. Yet in public policy, the literature is more despondent about the ability of lessons from crises to successfully travel. This is the issue that Dr Maria Maley and I explore in Robodebt and the limits of learning: exploring meaning-making after a crisis
Leaders provide a frame through which the crisis can be understood, and to suggest a way of thinking – “pointing to a way forward out of the rubble”. This framing shapes what lessons can be learnt, if any. Exploring these framing processes helps explain why some organisations learn after crisis but others do not.
I don’t need to repeat for this audience the sorry story of Robodebt. At the heart of our research was a belief that the cultural failures evident in Robodebt were not confined to a couple of departments, but were actually evident across the APS (and across other Westminster-based public services). And we were concerned that the cultural failings identified by the Royal Commission may end up being ignored by other parts of the public service.
Peeking behind the internal communications curtain
Our research relied on FOI requests to all 113 Australian Government entities with staff employed under the Public Service Act 1999 (PS Act) asking for internal communications about Robodebt. This approach gave us access to primary material, written for an internal audience, and allowed us to compare these with the ‘official discourse’ of the public statements by the APS leadership. This is also a way to identify inaction, which is often methodologically difficult – where agencies advised us that “no such document exists,” it is not unreasonable to conclude that the relevant event did not occur, within the timeframe of the FOI request (which was the six-month period following the release of the Royal Commission’s Report).
There is a constant tension, in both the academic and practitioner literature about the impact FOI has on the willingness of public servants to put their advice in writing and provide the necessary frank and fearless advice. And I have received informal criticism about this research, suggesting it may further discourage putting things in writing. So, FOI is no research silver-bullet, but it’s should be considered as a vital arrow in a researcher’s quiver – to mix my metaphors. We thought this was an appropriate approach in this case particularly because the Robodebt Royal Commission specifically addressed this issue of record keeping failures, and a tendency to deliberately fail to record key decisions and discussions.
Who tried to learn lessons?
So, how did APS agencies go? Our conclusion was based on a simple premise – if leaders do not talk about lessons, then no lessons can be learnt. Out of the 113 APS agencies, 50 did not communicate with their staff about the lessons of Robodebt in the immediate aftermath of the Royal Commission. This represents more than 45,000 public servants, which is over 25 per cent of the public service. This absence of communication still sends an implicit message: there are no lessons to be learnt, it is not relevant for us.
However, some leaders responded with genuine introspection and engaged with the challenge to public servants’ sense of identity that Robodebt represented. The head of the National Library of Australia, Dr Marie-Louise Ayres, an agency with no connection to Robodebt, reflected in an all-staff presentation on 4 August 2023:
I want to ask myself – can I set aside my own assumptions about how we conduct ourselves in our organization, recognize that I’m now a long way from the coalface, and truly listen if an allegation, or even a suggestion of something really disturbing happening comes my way. Can I create an atmosphere … so that serious matters are taken seriously? [Original italics]
At an Australian Bureau of Statistics SES planning day, Australian Statistician David Gruen emphasised to his senior staff the importance of creating a culture where ‘people feel supported if and when they seek to raise difficult issues with their colleagues or superiors’. The Bureau of Meteorology emphasised that ‘the public administration shortcomings … could occur, without any malicious intent, in any government agency.’ Similar discussions about the meaning of the failures of Robodebt happened at many other agencies. These are clearly positive signs that some leaders understood the need for cultural reflection, even when their organisations weren’t directly implicated.
It’s important to recognise what our research did not cover. We focussed solely on the period between 1 July 2023 and 31 December 2023. We have no knowledge of what happened after that. It is possible that other agencies took extensive action in early 2024 – which we would not have uncovered. Our analysis was dependent on the material released by agencies following our FOI request.
Where to next?
Since we finalised our research, there have been mixed messages coming out of the APS. In September 2024, (after the time period covered by our research), the APSC released the Final Report of the Centralised Code of Conduct Inquiry Taskforce. Beyond the findings about 16 individuals, the report identified “interconnected matters that may indicate systemic issues of broader applicability.” I think that this report is a better reflection of the underlying issues than the APS Integrity Taskforce’s official report Louder than Words (which studiously ignored the actual underlying problems), put out by the APS Secretaries Board. The Taskforce report emphasised that “[u]ltimately it is about the culture and leadership of the agencies.” It also directly addressing the issue of over-responsiveness:
“an apparent culture of prioritising delivery of Government objectives at all costs created a reluctance to identify risks… the Taskforce observed examples of poor behaviour by senior leaders being overlooked in pursuit of achieving results. These leaders were often identified as ‘deliverers’, people who could be counted on to deliver on Government priorities with little regard for the costs to staff wellbeing, workplace culture and good public policy.”
As a former public servant, who worked for some of these “deliverers,” this is a welcome recognition, and I hope this message is heard and repeated throughout the APS.
However, more recent media reports seem to indicate that some senior public servants continue to deny that the lessons of Robodebt are relevant for them, with a senior pubic servant reflecting, anonymously, to a journalist that:
secretaries got that robodebt had to be dealt with strongly, but some felt that the unlawful scheme had occurred in a couple of agencies, and led to the whole public service being tarnished.
To quote Jeff Daniels in The Newsroom, "the first step in solving any problem is recognizing there is one." This is leadership. The importance of individual leaders, with their own styles and approaches, for organisational learning. What individual leaders pay attention to, and do not pay attention to, sends powerful messages to their organisations.
Learning from others’ failure can often be much harder than learning from your own failure. As Dr Ayres, the head of the National Library of Australia said to staff, “the feeling around Robodebt is a combination of horror, and grief, a good bit of ‘there but for the Grace of God go I’”. In that moment, leaders face a choice. Some, like Ayres, chose to face the uncomfortable and challenging issues raised by Robodebt head on. Others pulled back, concerned that engaging would imply fault or failure in their own operations – that the underlying causes couldn’t possibly be present in their organisations.
But denying the problem is not the same as being immune from the problem. As the Bureau of Meteorology rightly noted, such failures could occur in any agency. Whether they do or not depends, in no small part, on whether leaders are willing to firstly face difficult truths, even – or especially – when they’re not our own, and secondly honestly and openly communicate with their staff, and help them understand and learn from such serious failures.
Dr Daniel Casey is a lecturer in politics and international relations at the Australian Catholic University and visiting fellow at the Australian National University (ANU).
Moderator: Dr Jeremiah Brown