Administrative law and cultural change

John McMillan, Australian Information Commissioner: Presentation to the Australian National University, Public Law Weekend, Canberra, 24 October 2014

Perspectives on administrative law

Administrative law can be viewed from many perspectives. A popular perspective is to analyse the reasoning in important judicial decisions, commonly of the High Court. Another is to study administrative law review by a particular tribunal or in a specific doctrinal or subject area. Yet another – and the perspective adopted in this paper – is to examine whether and how administrative law can review make a difference beyond settling an individual dispute.

A common theme in administrative law writing is that individual case rulings are important because they can have an effect beyond providing justice or a remedy to the applicant in that case. This theme is sometimes explicit, pointing for example to a new principle or safeguard in a ruling that can be relied upon by others. At other times it is a more generalised or unstated theme, expounding for example that a ruling will bolster respect for the rule of law.

Either way, the difficulty is that there is a high degree of assumption in connecting an individual ruling to a systemic result. How do we know that individual rulings become understood and respected throughout government? Most administrative decision making is not externally or independently reviewed or scrutinised. Nor, for the most part, is there any public record of when or how an administrative law ruling triggers a change in administrative practice.

Building a theory of how administrative law can cause systemic or cultural change may require a different approach. I will illustrate that point by exploring the role that non-judicial mechanisms can play in embedding administrative law values in government practice. I will draw from my own experience in two areas, as Commonwealth Ombudsman (2003-2010) and Australian Information Commissioner (2010-2014).

An avowed objective of both offices is to work at both the individual and the systemic level, by providing justice and a remedy to individual clients or complainants, while striving to improve government practice so that others may not need to approach the office. This dual objective was reflected in the tag line of each office during my term – of one, ‘helping people … improving government’; of the other, ‘protecting information rights … advancing information policy’.

The Ombudsman and cultural change

The central function of the Ombudsman is defined by statute as one of investigating administrative action, either upon receipt of a complaint or an own motion basis.[1] Lying behind that bare legislative description is a more powerful opportunity for the Ombudsman to use an individual error to highlight a larger or systemic problem that needs to be addressed. Individual cases can be used strategically to improve administration and trigger cultural change.

That point is well illustrated by the events that followed two notorious immigration incidents that arose during my term, concerning Ms Cornelia Rau and Ms Vivian Alvarez. I shall briefly recap the circumstances in each case.

Ms Rau was a foreign-born Australian resident who was found in a bewildered state in a public place in 2004. Police and immigration authorities formed the view that she was an unlawful non-citizen, and she was taken into immigration detention for a period of ten months until she was identified by her family from news reports.

The Australian Government established an independent inquiry led by former federal Police Commissioner, Mr Mick Palmer.[2] His report accepted that Ms Rau’s original apprehension and detention as a suspected unlawful non-citizen was lawful, but was highly critical of the Department of Immigration’s failure to move beyond its initial scanty assessment. Departmental failures identified by the Palmer report that led to Ms Rau’s continuing and unlawful detention were poor information flows, deficient factual evaluation, inadequate staff training, substandard mental health assessment, poor supervision of coercive action, a faulty cultural outlook, and defective management arrangements with service providers.

While Ms Rau’s case was under investigation another serious matter surfaced concerning the treatment of Ms Vivian Alvarez Solon. She was an Australian citizen who was admitted to a hospital psychiatric unit in 2001 after being found injured in a public space. After three months she was taken from the hospital into immigration detention as a suspected unlawful non-citizen, and a week later removed to her country of birth, the Philippines.

Doubts about her removal surfaced during 2003-05, when inquiries were made by a Queensland Police Missing Persons unit, her disappearance was featured on a television program, and her former husband made efforts to locate her. Department of Immigration officials became aware during this period that a serious error had occurred in deporting an Australian citizen, but no action was taken. It was not until Ms Alvarez’s former husband contacted a Ministerial hotline that an internal inquiry was established, and soon after an external inquiry headed by a former State Police Commissioner, Mr Neil Comrie.

Before that inquiry was completed the inquiry team led by Mr Comrie was transferred to the Commonwealth Ombudsman’s office, so that the inquiry could be completed under the Ombudsman Act 1976.[3] The report in 2005 echoed themes in the Palmer report about Departmental failures at every stage of Ms Alvarez’s apprehension and detention. These occurred in applying the law, evaluating information, following up doubts, dealing with mental health problems, resolving problems and in Departmental culture.

It was shocking enough that two incidents of this gravity could occur in a national government department. Equally worrying was that they could be the proverbial tip of an institutional iceberg. They showed that Departmental culture did not prevent mistakes of this kind after more than two decades of intensive scrutiny through judicial review, tribunal review, Ombudsman and Human Rights office investigations, Parliamentary oversight and media scrutiny.

The Government accepted that a change was required, in systems, procedures, culture, values, people and scrutiny. Among the wholesale changes within the Department were a new leadership team, an agency restructure, new internal governance arrangements for supervision and reporting, new information technology and case management systems, new procedures for verifying information and managing difficult cases, new internal audit arrangements, new staff training and a cultural transformation program.

Government accepted that enhanced external scrutiny by the Ombudsman was another essential reform. This was along four lines. Firstly, the Ombudsman Act was amended to give the office an additional and specialist designation of Immigration Ombudsman.[4] This was accompanied by new functions (described below) and additional funding. The new Ombudsman title conveyed that a general oversight body with across-government experience in applying principles of good administration, could adapt that experience to a specific government zone that threw up special problems that required a tailored response.

An allied legislative reform that conveyed a similar message was the extension of the Ombudsman’s jurisdiction to the actions of contracted service providers.[5] This was to enable effective oversight of how detention centres are managed and health and other services are provided. More generally, this was a landmark development in breaking down the public/private divide in a specific area of administrative law oversight.

A second reform was that 247 cases of possible wrongful detention were referred to the Ombudsman for individual investigation. Nine reports were published. Two dealt with individual cases, while another six consolidated reports addressed common themes in the 247 cases of children in detention, mental health and incapacity, data problems, notification problems, detention process issues, and other legal issues. Those eight reports identified legal and factual errors in nearly all 247 cases, including the wrongful detention of 26 Australian citizens, and wrongful detention of two people for over five and six years.

The exercise culminated in a ninth report, Lessons for Public Administration.[6] The report spelt out ten lessons that should be heeded by all of government. The same message in all ten lessons was that poor and injurious decision making can stem from mistakes made by individual decision makers, but stems more commonly from defective administrative systems and a flawed administrative culture. These include poor record keeping, administrative drift, erroneous assumptions, inadequate supervision of coercive powers, failure to escalate unresolved and difficult cases, improper reliance on information technology systems, ineffective internal communication, and not learning from mistakes.

A third reform was a new Ombudsman function of preparing a report to be tabled in the Parliament on each person held in immigration detention for more than two years.[7] In the first three years of that function, when 436 reports were prepared, the number of people in long term detention dropped from 149 to 34. The function was accordingly supplemented by a new practice of preparing an initial assessment after six months detention. [8]

The fourth area of reform, drawing on the additional budget funding for the Immigration Ombudsman role, was more intensive oversight of immigration administration generally. This included more regular inspection of immigration detention facilities, unannounced visits to detention centres, monitoring compliance and removal operations, and preparing own motion reports on topics as varied as immigration debt waiver, visa cancellation, criminal deportation, notification of decisions, complaint handling and freedom of information delays.

It is difficult to measure or quantify the objective success of those reforms, or to demonstrate how they contributed to cultural change in the immigration program. The immense scale of Departmental operations is one reason for that difficulty. At the time, for example, the Department employed over 8000 people in offices around Australia and the globe, who annually managed more than 26 million passenger movements, granted more than 8 million visas, located over 11,000 unlawful non-citizens, detained nearly 4000 people, and removed another 7000.[9] Those figures, particularly on compliance and detention activities, can change dramatically over time. By 2013, for example, the Ombudsman was back to preparing 709 reports that year on people detained for more than two years.[10]

What was observable, however, was the Department’s avowed commitment to cultural and business transformation, and its reliance on Ombudsman oversight as integral to that process. For a number of years after 2005 the Secretary’s overview in the Department’s annual report contained a dedicated section on how the Department was responding to the Rau and Alvarez inquiries and Ombudsman reports. For example, in the 2007-08 annual report the Secretary wrote that:

The department has worked closely with the Ombudsman and his office, and other accountability bodies and stakeholders over the past three years to ensure major changes and improvements are made to our systems and processes.

As to the Ombudsman reports on the 247 referred cases, the Secretary wrote that:

These assessments are guiding the department’s remedial action in relation to internal systems, records changes and the offering of compensation and other remedies to the affected people. [This] demonstrates the strength of the department’s commitment to remedying past wrongs suffered by individuals.[11]

An independent review by Ms Elizabeth Proust commissioned by the Department in 2008 reached similar findings about the Department’s substantial reform progress.[12] This was also observed at a transactional level by Ombudsman staff. Our experience was that the Department was more open to scrutiny; it responded more constructively to requests; it welcomed external participation in internal review and training activities; there was greater confidence in the Ombudsman’s office that the unchecked mismanagement exposed in the Rau and Alvarez cases would more easily be detected and hopefully not re-occur; and, anecdotally, the Department Secretary, Andrew Metcalfe, became a vigorous promoter across government of the Ombudsman’s Ten Lessons report.

Broader lessons about cultural change that can be drawn from that immigration case study are taken up later in this paper.

The OAIC and FOI cultural change

The enactment and operation of the Freedom of Information Act 1982 in Australia is another illustrative case study in cultural change.

Cultural temperament lies at the heart of FOI. The FOI Act was enacted in 1982 against the backdrop of an entrenched tradition of government secrecy. At the time, government had an unreviewable discretion to decide when to release information; the statute book contained stiff criminal and disciplinary penalties for unauthorised release of government information; and legal doctrine supported the notion that government information was the property of the Crown and that ‘The counsels of the Crown are secret’.

The FOI Act reversed those features. It created an enforceable legal right of access to government documents; the right was exercisable by all members of the public on equal terms; an agency or minister could refuse access only on an exemption ground in the Act; the agency’s or minister’s decision was externally reviewable; and the onus of justifying a decision to refuse access rested with the agency or minister.

The FOI Act supported a new era of transparent government in Australia. It became standard practice that individuals could readily obtain access to their own case files; far more information was released by government in the form of reports, submissions and meeting papers; agency officials shed their anonymity and participated actively in public affairs; and, as technology transformed government, agencies utilised the web and online mechanisms to proactively publish even more information and to make service delivery more interactive and responsive.

But there is always an undertow. Transparency can be inconvenient. An FOI request can take time to process and be a wasteful diversion from other tasks, at times requiring inconsequential documents and email chains to be located and provided to requesters who may do nothing with the material. Disclosure can be awkward – the public release of documents prepared for a narrow or specific internal purpose can sidetrack an agency into a public debate that misrepresents the documentary context. The FOI Act also enables fishing expeditions that are driven not by a need to know but by the chance of crafting a news story or fomenting a public scandal. There is truth in the lament of former Prime Minister, Tony Blair, who introduced FOI in Britain, that ‘For political leaders, it’s like saying to someone who is hitting you over the head with a stick, “Hey, try this instead”, and handing them a mallet’.[13]

Open government is entwined with culture, both political and bureaucratic. Senator Joe Ludwig, who introduced wide-ranging reforms into the Parliament in 2010 to update the Freedom of Information Act 1982 and to create the Office of the Australian Information Commissioner (OAIC), aptly commented that the objective was ‘to switch the mindset … from information control to information sharing’.[14]

How was that to be done? A standard way is to construe the new criteria in the FOI Act in a manner consistent with a presumption of open government. A key element of the 2010 reforms was that the OAIC could undertake merit review of agency and Ministerial decisions to refuse access (called Information Commissioner review, or IC review). In the first three and a half years of the new changes the OAIC resolved 1345 review applications and published reasons for decision in over 200 cases.

Individual review decisions can change decision making practice, but will not alone transform government culture in processing annually upward of 28,000 FOI requests and dealing with information release issues in many other situations.[15] Indeed, many IC review decisions confirm an agency’s exemption claim for the simple reason that a free right of appeal is often exercised in hopeless cases. That has exposed the OAIC to the (in my view, baseless) criticism that its decisions often condone government secrecy.

From the outset the OAIC relied strongly on its other functions to promote a more enlightened open government culture. One step was to promote key messages that are both memorable and would usher in a new era of open government. They were that ‘government information is a national resource to be used for public purposes’; government information is better described as ‘public sector information’; and access to information laws embody a ‘presumption in favour of disclosure’. Those phrases are now widely referred to and understood across government

The OAIC banner, in which the three corporate colours are entwined to convey symbolically that the FOI, privacy and information policy functions must be integrated in a coherent scheme of responsible information management, was another way of conveying a new philosophy to government.

Another step was to make it easier for agencies to make the right decision – or, conversely, harder to make a bad decision that would go unchecked. We published extensive guidelines on the FOI Act, and numerous guidance documents on technical issues such as processing times, calculating charges, redaction, website publication, disclosure logs and sample letters. These are among the most commonly accessed materials on the OAIC website, and are frequently referred to in the reasons for decision of agency decision makers.

The next step was to engage directly with agency staff in promoting open government. This was done at all levels – at meetings with and presentations to senior agency managers when the reforms were introduced; through an Information Contact Officers Network that provided ‘news alerts’ and held meetings that are regularly attended by upward of 130 FOI and privacy officers; by a telephone advice service that answers hundreds of calls annually; and through telephone discussion and preliminary case analysis of new review applications.

Informal interaction and dialogue between a regulator and other parties provides a valuable opportunity to work through the practical impediments and attitudinal barriers to implementing the rules and values of a legislative scheme. The discussion can also underscore the limitations of the legislative scheme. In this instance, it is that FOI procedures are too easily a cumbersome and combative pathway for obtaining access to government information.

The FOI Act is an essential legislative anchor for open government. But in a changing world where most information is digitised and people go online to contact government, FOI must be supplemented by other schemes. Two that the OAIC actively promotes are administrative access and proactive agency disclosure. That is reinforced in a widely-known OAIC publication, the Principles on Open Public Sector Information.

Has this work been successful, and will the benefits be lasting? The second question arises because there is legislation presently before the Parliament to disband the OAIC at the end of this year as a government streamlining and budgetary measure.[16] Other oversight mechanisms will take the place of the OAIC. However, as I comment later, a key requirement for cultural change is persistent reinforcement of the cultural messages. So the baton passes to others.

As to the first question, anecdotal signs point strongly to the cultural change message taking root. Government agency disclosure logs that publish documents released under the FOI Act include many that were far harder to obtain in past years, such as ministerial submissions, hot issues briefs, expense accounts, audit reports, correspondence, submissions and meeting papers. FOI is now front-of-mind for senior executive officers. They acknowledge increasingly that it can be less troublesome to release than to withhold. ‘Write for the world to see’, is the advice given by Australian Taxation Office managers to all staff.

There is also greater consistency in FOI administration across government, which was a major criticism prior to 2010.[17] A comforting statistic in 2013-14 is that while agencies received 14% more FOI requests than in the previous year, the reported cost of FOI Act compliance reduced by 7%. It costs less to release than to withhold.

Cultural change lessons

From those and other case studies, can we draw general lessons about administrative law and cultural change? I will finish with four observations.

The first step is to identify the values that we expect should be deeply embedded and respected in government administration. There is no definitive list of administrative values, but those commonly mentioned include legality, rationality, fairness, consistency, transparency, answerability and respect for individual rights.

Different strategies may be needed for advancing each of those values. Judicial review, for example, is especially effective in spelling out the meaning of legality and, in recent jurisprudence, concepts of rationality, reasonableness and justification.[18] But other methods – such as auditing, own motion investigations and record inspection programs – can be more effective in identifying whether administrative law values are routinely respected and applied in administrative decision making.

Cultural deficiency in an agency generally goes hand-in-hand with system inadequacy. Common examples are faulty record keeping, unskilled decision making, administrative drift, poor internal information flows, inadequate quality control and follow-up, and IT systems that are not aligned with the business processes of the agency.

Individual examples of those problems can be highlighted in all mechanisms of administrative law oversight. Whether the individual example is an illustration of a deeper administrative inadequacy or malaise requires different analysis. Interestingly, the Auditor-General, through performance auditing, has led the way in demonstrating this point. The Auditor’s work can provide salutary lessons for the administrative law community.

A second lesson is that individual cases can provide a powerful illustration of systemic failings and trigger a process of general reform. The spectacle of injustice suffered by Cornelia Rau and Vivian Alvarez had a greater impact in prompting comprehensive immigration reform than other processes in the preceding two decades.

There are many similar examples in government of how publicised bungles or errors can galvanise the need for a general review. ‘If it happened once it can happen again’ is the motivation for change. Individual cases can provide a window to a larger world. An illustration this week was that a brief case study in the annual report of the Inspector-General of Intelligence and Security about misuse of a firearm by a special forces soldier was widely publicised and followed by a Defence announcement that a general review will be conducted.[19]

From a cultural change perspective, the crucial issue is how to establish a link between the individual problem and a process of reform. Change will occur only if there is a political or executive preparedness to take that step. This again is where a regulator, such as an Ombudsman or Commissioner, can play a special role in applying persuasive pressure for further action. This can be done in recommendations, reports, parliamentary submissions, meetings, press releases or, when those don’t work, persistent nagging or shaming.

A third lesson is that cultural change is not necessarily permanent. In administrative systems there can be a default position that will re-emerge if the opportunity arises. An example I gave earlier is the predisposition of ministers and public servants to information control rather than information release.

Sustained pressure and periodic monitoring may be required. Complaints can be another high-volume indicator of recurring problems that need a permanent fix. Requiring agencies to provide a report or self-assessment on how they have implemented an adverse finding is another technique.

The fourth and final lesson is that the administrative law system can be a more effective agent for elevating the standards and culture of administration if the work of the administrative law review bodies can be integrated and harmonised. The need for a coherent, comprehensive and integrated system of Australian administrative law was indeed the principal objective of the Kerr Committee in its foundational report in 1971.[20]

Yet that objective, more than any other proposal or objective of the Kerr Committee, has languished. The jurisprudence of courts, tribunals and Ombudsman is relatively contained. Only infrequently will a court refer to or adopt the insights of the tribunal whose decision it is reviewing. My recent experience is that most Administrative Appeals Tribunal members pay the same compliment to my office in FOI review cases.[21] Similarly, I have often heard judges and tribunal members comment that they witness egregious administrative mistakes, but only once during my seven years as Ombudsman did a tribunal member refer a matter to me for possible own motion investigation.

There is, in short, an unhealthy separation, comparison and ranking of administrative law review agencies. It is an attitude that has deep roots in legal culture and in administrative law teaching in Australia. I have on other occasions criticised academic evaluation of the different mechanisms of administrative law review, using time-worn stereotypes that are based on outdated notions of independence, power and effectiveness.[22]

This narrow perspective of the comparative effectiveness of the different review mechanisms resurfaced this week in the wake of the public prominence of the office of Inspector-General of Intelligence and Security. The office is to be given extra funding and a stronger role to oversight the exercise of new counter-terrorism powers by security intelligence agencies. One academic colleague doubted the effectiveness of the office as its funding had not kept up with the enormous increase in personnel in the intelligence community.[23] When is it said that courts are ineffective because they have not grown at the same rate as the Australian government and business community? Another commentator said the office of Inspector General needed to be restructured, as a witness may have lied to it in a publicised investigation.[24] When is it said that a court or tribunal system should be restructured because a witness gave perjured evidence? And another colleague observed that the office is merely a ‘paper tiger’.[25] It is a pity, nearly forty years through the life of the Australian administrative law system, that this simplistic and hoary metaphor is still bandied about.

One other endeavour in recent literature to reposition the Ombudsman and similar bodies in legal doctrine as a possible fourth or integrity branch of government, was slapped down by former Justice Gummow and Western Australian Chief Justice Martin.[26] They firmly rejected the idea that we need to stimulate debate about the continued suitability of constitutional models, to take account of the profound changes in the dispute resolution landscape that have evolved over the past forty years.

I finish with the brusque observation that discussion of administrative law and cultural change in Australia may need a broader focus.