Federal Law Review
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Federal Law Review |
|
Chris Finn[*]
Australians, it is said, live in a representative democracy. As citizens of that democracy they are ultimately able to hold their political representatives to account via the electoral process. In the interim between electoral episodes they form the views upon which that holding to account will be based. At the same time, they participate in and evaluate governmental processes in a variety of ways, some of them statutory, some of them via the Parliament of the day, and many of them via the media, lobby organisations, interest groups and other proxies.
Informational transparency is central to this holding to account. This was famously recognised by the High Court in the so called 'freedom of political communication' cases in 1992. Those cases constituted the first explicit judicial recognition that 'the sovereign power which resides in the people [and] is exercised on their behalf by their representatives'[1] was more than simply a political doctrine, but was in fact a constitutional fundamental capable of limiting, as it did in those cases, Commonwealth legislative power.
The High Court expressly commented on the centrality of informational transparency to any genuine exercise of electoral choice. McHugh J stated that: [2]
The electors must be able to ascertain and examine the performances of their elected representatives and the capabilities and policies of all candidates for election. Before they can cast an effective vote at election time, they must have access to the information, ideas and arguments which are necessary to make an informed judgment as to how they have been governed and as to what policies are in the interests of themselves, their communities and the nation. … Only by the spread of information, opinions and arguments can electors make an effective and responsible choice
Mason CJ observed that:[3]
Absent such a freedom of communication, representative government would fail to achieve its purpose, namely, government by the people through their elected representatives; government would cease to be responsive to the needs and wishes of the people and, in that sense, would cease to be truly representative.
In short, any real electoral choice must be an informed choice.
The Freedom of Information Act 1982 (Cth) ('the Act') came into operation a full decade prior to these statements of the High Court. But the parallels with its twin rationales of democratic accountability and citizen participation are striking.[4]
It is doubtful, however, that these high ideals have been met. Rather, both Commonwealth and State freedom of information ('FOI') Acts have been subject to sustained criticism for the inadequate informational access rights that they provide.[5] In part, this is due to the broad ranging exemptions to access contained within the Acts; in part it is also due to the conservative interpretations of those exemptions given by individual FOI decision makers in the first instance and by tribunals and courts upon review. Executive preferences for, and agency cultures of, secrecy and non-disclosure have remained largely unaltered by 25 years of FOI legislation, and those preferences and cultures have gone largely unchallenged by the decisions of courts and tribunals. In the result, the public's 'right to know' has become little more than a cipher. While individuals are generally able to access their own personal files successfully under FOI legislation, broader 'public interest' requests lodged by newspapers, opposition politicians and by lobby groups face enormous difficulties.[6] Large cost burdens are imposed by agencies upon such requestors, broad exemptions are claimed for key documents of interest and the use of conclusive certificates allows Ministers or agencies reluctant to release documents to pre-emptively determine the question of 'public interest' against any such release.
'Public interest' is a key concept in FOI Acts. However, it is a complex and multi-faceted concept, which is implemented in a variety of ways in the various exemption provisions and which is subject to two key tensions within these Acts. First, there is the question of the substantive content of 'the public interest' in relation to FOI release. While FOI advocates draw upon the 'right' of access conferred by s 3 of the Act and stress the fundamental democratic interest in access to information, and hence the release of sought after documents, to further political discussion, it is evident that the statutory rights to access information actually conferred by FOI Acts are of a very qualified kind. In particular, the various exemption provisions expressly qualify and limit those access rights. These exemptions are predicated upon assumptions that release of documents will not always be in the public interest, democratic accountability notwithstanding. Taken together, they constitute an extensive 'shopping list' upon which a Minister or agency can draw if they are minded to refuse release.
There is perhaps an even more fundamental tension in relation to the 'public interest'. This is simply the question of who gets to decide what is in the public interest. While FOI Acts provide a broad legislative framework, the legislation in itself is usually not determinative. Rather, the public interest for or against release in response to a particular request for documents has to be evaluated and decided, initially at least, by an agency or Minister who could well be viewed as an interested party. The release of documents which may suggest that maladministration or worse has occurred, or that unpopular policies are being considered by the government of the day, is highly likely to be resisted.
Given the less than impartial nature of such decision making, independent review of 'public interest' determinations is critical to the efficacy of FOI legislation. As successive Commonwealth governments have ignored the recommendation that they create an independent FOI Commissioner to adjudicate upon such matters,[7] it is left to the Administrative Appeals Tribunal ('the AAT') and then the courts to make these determinations.
However, conclusive certificates remove the possibility of such independent adjudication of the public interest. Rather, they allow the interested Minister or agency to 'conclusively' determine that the public interest favours non-disclosure of the sought after documents. The result is that the independent review role of the AAT and the courts is reduced to one of confirming that the person issuing the relevant certificate had 'reasonable grounds' for their belief as to the balance of the public interest. The outcome is that conclusive certificates are seldom overturned and the role of determining the public interest is effectively retained by the Executive.
The use of conclusive certificates has encountered particular criticism in relation to s 36, the 'internal working documents' exemption in the Act.[8] The Commonwealth Act is unusual in providing for conclusive certificates in relation to this type of exemption. No State FOI Act provides for conclusive certificates in relation to such documents, and the other exemption provisions for which they are available relate to matters such as Cabinet and Executive Council documents, national security and relations between governments, where the arguments against exemption are in general much stronger and the facts are much more likely to be peculiarly within the knowledge of the Executive.
Despite this, there is a long history of issuing conclusive certificates stating that the 'public interest' militates against release of internal working documents.[9] In perhaps the best known of these cases, Re Howard,[10] the then Treasurer, Paul Keating, had issued a conclusive certificate blocking the release of documents sought by the then Deputy Leader of the Opposition, John Howard. The certificate was upheld on review by the AAT. Many of the 'public interest' grounds for the issue of a conclusive certificate in that case (the so called 'Howard factors') were to reappear in McKinnon's case. The litigation thus provided the High Court with an opportunity to clarify the nature of the public interest in relation to claims of exemption under s 36, to assert a role for the courts in determining that public interest, and to interpret the Act in the light of its role as part of the structure of a representative democracy. It was an opportunity the majority of the Court chose to avoid.
Michael McKinnon is the FOI editor of 'The Australian' newspaper. In October of 2002, he made applications under the Act for the release of material relating to 'bracket creep' in the Commonwealth income taxation system. After some refining, this became a request for:[11]
Reports, reviews or evaluations completed in the 12 months from 3 December 2001 to 3 December 2002 detailing the extent and impact of bracket creep and its impact on revenue collection of income tax, including information in relation to higher tax burdens faced by Australians and/or projections of revenue collection increases from bracket creep, but excluding documents that have already been released publicly or duplicate copies of documents.
The application was originally made to the Australian Taxation Office, but subsequently transferred to the Commonwealth Department of Treasury.
In December of 2002, McKinnon made a further application to the Treasury, this time for documents relating to the First Home Owners Scheme. In particular, the application sought release of:[12]
Documents relating to any review/report or evaluation completed on the First Home [Owners] Scheme in the last two years, including documents summarising the level of fraud associated with the program, its use by high wealth individuals and its impact on the housing sector's performance in the Australian economy.
Both applications met with almost complete refusal. While some documents were released during the course of the Tribunal proceedings, 36 'Bracket Creep' documents and 11 relating to the First Home Owners request remained unreleased when the Tribunal made its decision. Those refusals were made on the basis that the documents in question were 'internal working documents' as defined in s 36 of the Act, and that their release would be contrary to the public interest.
Section 36(1) provides:
Subject to this section, a document is an exempt document if it is a document the disclosure of which under this Act:
(a) would disclose matter in the nature of, or relating to, opinion, advice or recommendation obtained, prepared or recorded, or consultation or deliberation that has taken place, in the course of, or for the purposes of, the deliberative processes involved in the functions of an agency or Minister or of the Government of the Commonwealth; and
(b) would be contrary to the public interest.
As the wording of this provision makes clear, the requirements of both s 36(1)(a) and s 36(1)(b) must be met in order for the exemption under s 36 to apply.
McKinnon applied to the AAT for review of the decisions refusing access to the documents. Shortly before the matters were listed for hearing, the Commonwealth Treasurer, acting under s 36(3) of the Act, issued conclusive certificates stating that release of the documents would be contrary to the public interest and listing seven grounds for this conclusion.
Those seven grounds raised two main types of claims: first, that disclosure would compromise confidentiality within government (grounds (a), (b) and (f)), and second, that disclosure would be likely to mislead the public (grounds (c), (d), (e) and (g)).
Section 36(3) provides as follows:
Where a Minister is satisfied, in relation to a document to which paragraph (1)(a) applies, that the disclosure of the document would be contrary to the public interest, he or she may sign a certificate to that effect (specifying the ground of public interest in relation to which the certificate is given) and, subject to the operation of Part VI, such a certificate, so long as it remains in force, establishes conclusively that the disclosure of that document would be contrary to the public interest.
The effect of a conclusive certificate is to greatly limit the role of the AAT upon review of a decision to refuse release. Rather than reviewing the decision 'on the merits' including the claim that release would be contrary to the public interest, and making its own determination of that question, the role of the AAT on review is limited to that set out in s 58(5) of the Act, which provides that:
Where application is or has been made to the Tribunal for the review of a decision refusing to grant access to a document in accordance with a request, being a document that is claimed to be an exempt document under section 36 and in respect of which a certificate is in force under that section, the Tribunal shall, in a case where it is satisfied that the document is a document to which paragraph 36(1)(a) applies, if the applicant so requests, determine the question whether there exist reasonable grounds for the claim that the disclosure of the document would be contrary to the public interest.
Thus, under s 58(5), the AAT does not determine whether release of the documents would be contrary to the public interest, but only whether there are 'reasonable grounds' for believing this. Downes J, President of the AAT, heard the matter in 2004. His decision, released in December 2004, concluded that there were indeed 'reasonable grounds' for the issue of the certificates in question.[13] Thus, the release of the documents in question was conclusively contrary to the public interest and McKinnon's application for review failed.
McKinnon appealed to the Full Federal Court. In August of 2005, that court held by majority that Downes J had not erred in his interpretation and application of s 58(5).[14] McKinnon appealed further from that decision to the High Court. In September 2006 the High Court, again by majority, upheld the decision of the Full Court.[15]
The High Court's task was to determine whether the AAT had correctly performed its statutory role in conducting the limited form of review provided for by s 58(5). As Hayne J noted in the High Court, this essentially raised two issues.[16] First, what was the exact legal requirement imposed upon the AAT by s 58(5)? What was the precise nature of the statutory task it was required to undertake when reviewing the issue of a conclusive certificate? In particular, did s 58(5) of the Act require the Tribunal to take into account and balance public interest considerations favouring disclosure of a document when determining whether reasonable grounds existed for a claim that disclosure would be contrary to the public interest? Second, what exactly had Downes J done? Had he, or had he not, fulfilled the legal requirement? The Court divided, in different ways, on these two issues.
It is clear that s 58(5) allows the AAT only a limited review power in relation to the decision to issue a conclusive certificate. Review under s 58(5) is quite unlike a typical AAT review of the exercise of discretionary power, which involves 'full merits review', ie, a complete re-exercise by the Tribunal of the discretionary power in question. Several members of the Court emphasised the atypical nature of the AAT's review function under s 58(5).[17] Similar observations had been made by both the Full Federal Court and by Downes J in the AAT.[18] Clearly, these statements are correct and the task of the AAT, whatever its precise nature, is not to perform a merits review when acting under s 58(5). It is the 'reasonableness' of the Minister's view that the release of documents would be contrary to the public interest which must be assessed, not the merits of that view.
However, both in the lower courts and the High Court, two distinct views emerged as to the exact nature of the AAT's task in determining whether there were 'reasonable grounds' for the issue of a conclusive certificate, as required by s 58(5). These may be termed 'global' reasonableness and 'stand alone' reasonableness.
The difference between these two approaches lies in the way in which they require the task of assessing 'reasonableness' to be undertaken. On the 'stand alone' view, it is sufficient for a subjectively held opinion to be reasonable if there is a single non absurd or rational ground which supports that opinion.
However, on the 'global' approach to the assessment of reasonableness, it is not sufficient to assess the rationality of one argument or opinion in isolation. Rather, that assessment can only be made after all relevant materials have been considered. Thus, there is the possibility that an opinion which, considered in isolation, might appear to support the reasonableness of the view that disclosure would be contrary to the public interest, might nevertheless fail to support the reasonableness of that view when considered in the context of all other relevant material.
In the Full Federal Court, the majority judgments of Tamberlin J and Jacobson J had adopted the 'stand alone' approach to the assessment of reasonableness. Jacobson J delivered the leading judgment. His Honour commented that:[19]
It may be accepted that a determination of where the public interest lies requires a balancing of competing factors. But s 58(5) requires the Tribunal to determine whether there are reasonable grounds for the claim that the disclosure would be contrary to the public interest.
In rejecting the broader 'global' approach, it was evident that Jacobson J was concerned that review by the Tribunal under s 58(5) did not become a form of disguised merits review of the Minister's decision to issue a conclusive certificate. In his Honour's view, such an approach:[20]
would negate the reasonable grounds concept and permit the Tribunal, through the back door, to come to its own opinion of what is in the public interest. That is not what s 58(5) requires.
Thus, Jacobson J preferred to follow the decision of Beazley J in Australian Doctors' Fund, which had also mandated a 'stand alone' approach to the assessment of 'reasonable grounds' for the purposes of review under s 58(5).[21]
Tamberlin J agreed with the reasons and orders proposed by Jacobson J. His Honour referred approvingly to the earlier Full Federal Court decision in Attorney-General v Cockroft which had held that 'reasonable grounds', in this context, denote grounds which are not 'irrational, absurd or ridiculous.'[22] Again, it was evident that his Honour was concerned that the AAT did not trespass into the 'merits' of the Minister's evaluation of the public interest with respect to the release of the documents in question:[23]
The question posed in s 58(5) is quite distinct from the question whether the non-disclosure is properly and finally determined to be in the public interest. It is important to focus on the consideration that the subsection is concerned with the existence of reasonable grounds for a claim and not with the answer that should be given to the question whether the non-disclosure was correctly considered to be in the public interest after a weighing of all the grounds and reasons. … In my view, s 58(5) of the FOI Act, in terms, is concerned with the threshold issue whether there is any non-absurd or rational ground for a claim that the disclosure of the material is contrary to the public interest…The existence of grounds for a reasonable claim that a conclusion should be reached addresses a different issue from whether the grounds, when finely balanced against all other relevant considerations, warrant the ultimate conclusion as to the public interest being reached.
After a discussion of the nature of the 'public interest, Tamberlin J re-iterated his view:[24]
Applying the above principles to the present case, one example of a facet of the public interest that is relevant is the desirability of preserving confidentiality of intra-governmental communications prior to making a decision. Another, and obviously competing, facet of the public interest is the desirability of transparency in public administration. If there is a ground that is not irrational, absurd or ridiculous for a claim that the first-mentioned facet of the public interest would not be served by disclosure, then that alone is sufficient to satisfy the requirements of s 58(5). It is not necessary in order to decide that limited question that the decision-maker should consider and weigh all the other facets, and the grounds which may reasonably support each of those facets, in order for s 58(5) to be satisfied.
This is a clear statement of the 'stand alone' approach. On this basis, Tamberlin J agreed with Jacobson J that the AAT had made no error of law.
In dissent, Conti J had adopted the 'global' approach to the assessment of reasonableness. His Honour referred to the unanimous High Court decision in George v Rockett,[25] which had observed, in the context of review of decisions to issue a search warrant that 'it must appear to the issuing justice, not merely to the person seeking the search warrant, that reasonable grounds for the relevant suspicion and belief exist'.[26]
Conti J reasoned by analogy that, in the s 58(5) context, the AAT was required to itself hold the requisite statutory opinion, ie, that 'reasonable grounds' existed for the belief that release of the documents would be contrary to the public interest. In order for the AAT to form that opinion, it must consider both the evidence supporting the reasonableness of that belief and also the evidence which disputed the reasonableness of that belief. These conflicting items of evidence must then be weighed or balanced in order for the Tribunal to reach its concluded view on the statutory question posed by s 58(5):[27]
I therefore think that the process of determination of the question 'whether there exist reasonable grounds for the claim for non-disclosure', in circumstances where the Tribunal is confronted with opposition to non-disclosure, requires the weighing and balancing of the grounds for non-disclosure as allegedly in the public interest with those grounds propounded by members of the public to the contrary, in order for the Tribunal to determine the reasonableness of those grounds for non-disclosure. The existence of 'reasonable grounds' is thus not to be resolved or identified in the vacuum of an isolated field of consideration articulated by a Minister, untested by reasonable postulations of views and opinions available to the contrary for the purpose of the inherently requisite balancing exercise. ... Rationality does not exist in a metaphysical vacuum, especially in the context of the elusive notion of 'the public interest', in the absence of a weighting exercise being undertaken by the decision-maker … Section 58(5) postulates the explicit notion of 'determin[ing] [a] question', not merely adopting one side or point of view where reasonable conflicting sides or views are propounded for similar resolution.
His Honour would have formally overruled the contrary decision of Beazley J in Australian Doctors Fund.[28]
The High Court was also to be divided over the precise nature of the s 58(5) task. For Gleeson CJ and Kirby J the task was a complex one which 'involve[d] an evaluation of the known facts, circumstances and considerations which may bear rationally upon the issue in question.'[29] Most significantly, to reach a conclusion as to whether there were indeed 'reasonable grounds' for the issue of a conclusive certificate it was necessary to consider all relevant matters bearing upon the public interest. Their Honours, reasoning by analogy, argued that individual facts or circumstances that, standing alone, might appear to constitute 'reasonable grounds', might nevertheless lose that character in the light of further facts or circumstances. Thus, for example, facts raising a reasonable suspicion that a person had committed a crime might lose that character in the face of a strong alibi:[30]
the question whether there are reasonable grounds for such a claim, or argument, or state of mind requires a consideration of all relevant matters and an assessment of the reasonableness of the claim, or argument, or state of mind having regard to all relevant considerations.
It is evident that their Honours adopted the 'global' approach to the AAT's task under s 58(5).
By contrast, Callinan and Heydon JJ held that the task of assessing the existence of 'reasonable grounds' for the ministerial opinion that release of documents would be contrary to the public interest was a much simpler task. It involved no balancing of arguments for and against the reasonableness of that opinion:[31]
The Act provides no mandate for any balancing exercise. To have regard to extraneous matters such as other competing reasons, if the requisite statutory reason for non-disclosure has been demonstrated, gives rise to a risk that a de facto balancing act will take place.
Thus, all that was required for the Minister to form the statutorily required state of mind was a single reasonable ground for the belief that disclosure would be contrary to the public interest:[32]
It does follow, as the majority in the Full Court effectively held, that if one reasonable ground for the claim of contrariety to the public interest exists, even though there may be reasonable grounds the other way, the conclusiveness will be beyond review.
Their Honours therefore can be seen to have adopted the 'stand alone' approach, whereby one 'reasonable ground', standing alone, is sufficient to meet the s 58(5) requirement, regardless of the existence or weight of evidence and opinions contradicting that ground.
This left the issue to turn upon the opinion of Hayne J. His Honour also adopted the 'global' approach, thereby agreeing with Gleeson CJ and Kirby J:[33]
The tribunal's task is not to be confined to examining those considerations separately. In particular, it is not to be confined to deciding whether one of the considerations advanced in support of a claim, that a document or documents should not be disclosed, can be seen to be based in reason. Rather, the tribunal's task is to decide whether the conclusion expressed in the certificate (that disclosure of particular documents would be contrary to the public interest) can be supported by logical arguments which, taken together, are reasonably open to be adopted and which, if adopted, would support the conclusion expressed in the certificate.
Thus, a majority of the High Court took the view that it was not sufficient for the AAT to find that there was at least one reasonable ground for the issue of a conclusive certificate. Rather, '[i]n deciding whether reasonable grounds exist for a claim, [the AAT] must take account of any relevant evidence that has been adduced and of any relevant arguments that have been advanced.'[34] The significance of this holding is discussed below.
Here again, the Court divided, albeit differently. Given the relaxed view they took of the s 58(5) requirements it was perhaps not surprising that Callinan and Heydon JJ held that the AAT had met them.
Gleeson CJ and Kirby J again held for the applicant, finding, although not without hesitation, that Downes J had erred in his approach to the s 58(5) task. In this, their Honours relied upon the assessment of that approach made by the Full Federal Court majority.
This left the result to turn upon the judgment of Hayne J. His Honour held that the applicant's submissions had not been made out. It had not been established that Downes J had failed to correctly carry out his statutory task under s 58(5).
Thus, by the narrowest of margins, McKinnon's application to the High Court failed.
It is a matter of some difficulty to determine the precise approach adopted by Downes J in hearing, and ultimately rejecting, the applications made by McKinnon. That difficulty is exacerbated by the provisions of s 58C of the Act which required that considerable portions of the AAT hearing be conducted in private. Some government evidence was in fact received by the AAT in the absence of the applicant and his representatives. Section 58C was also interpreted by his Honour as restricting his judgment to a quite limited account of both the disputed documents themselves and also the evidence which was given by witnesses for the Treasurer in support of the conclusive certificates. Thus, in assessing whether or not his Honour had in fact adopted the correct approach, both the Full Federal Court and subsequently the High Court were, to some extent, in the dark. This evidential difficulty makes it even less surprising that McKinnon ultimately failed to convince a majority of the High Court that the AAT had erred in its approach.
It is certainly evident that Downes J referred to the evidence of McKinnon's witnesses. Summaries of that evidence are given in his decision. In relation to the evidence of Mr Alan Rose, a former President of the Australian Law Reform Commission and a key witness for McKinnon, his Honour noted that this evidence was not the subject of cross examination and stated that 'I will accept his evidence.'[35] However, it is far from clear what was practically entailed by this 'acceptance', as his Honour's subsequent findings on the reasonableness of the public interest claims were in every instance contrary to the evidence of Mr Rose. Nor is it evident that his Honour engaged in the process of testing the Treasurer's claims against this evidence to establish their reasonableness, or otherwise, on a global basis. Somewhat cryptically, his Honour observed that 'to assess one expert opinion as definitive would not be to apply s 58(5).'[36] In relation to a key government witness, his Honour noted that his evidence 'supports the existence of an alternative reasonable opinion from the opinions expressed by the applicant's witnesses'.[37] That, it seemed, was sufficient to deny McKinnon any prospect of success.
His Honour's judgment essentially proceeded in three stages. First, he conducted a review of the authorities which disclosed a line of previous AAT decisions, admittedly not strictly binding, as well as some Federal Court authority evidencing previous acceptance of the 'reasonableness' of public interest claims similar to those made by the Treasurer on this occasion. Conflicting authorities were distinguished.[38] Downes J observed that '[w]here a document genuinely attracts consideration of a claim which has previously been upheld by [the AAT] under s 36 a finding that "there exist reasonable grounds for" a s 36 claim may not require a great deal more.'[39] His Honour then reviewed the evidence given by the parties, finding the evidence given by the Treasurer's witnesses to be supportive of the claims made in the conclusive certificates.[40] Finally, his Honour examined the documents themselves, concluding that they provided a 'substantial factual basis' for the public interest claims that had been made in the certificates.[41] No reference was made to the applicant's evidence at this stage.
On balance, though it remains a difficult matter to assess, it may well be that Downes J did not follow the global approach but rather adopted the 'stand alone' assessment of reasonableness which had been orthodox up to the High Court's subsequent pronouncement. If that is so, then with respect, Hayne J erred in deciding otherwise and should have decided for McKinnon rather than against him. However, as argued below, it is to be doubted that this would make any real difference to the eventual outcome of McKinnon's FOI application. Had the High Court decided differently and ruled that Downes J had erred, the matter would have been remitted to the AAT with appropriate directions for the subsequent re-hearing according to law. There, the narrowness of the distinction between the 'global' and 'stand alone' approaches, at least in the majority of cases, would have become starkly apparent. It is quite unlikely that any of McKinnon's witnesses, well qualified as they undoubtedly were, would have been able to conclusively refute the reasonableness of the view held by the Minister. Strong doubt might well have been cast, but that would not suffice to justify the AAT in finding that there were not 'reasonable grounds' for the ministerial belief underpinning the issue of the conclusive certificates.
The actual result in McKinnon (No 3) was obviously disappointing for FOI advocates and raises the question of whether the Act is in need of substantial reform.[42] It is worth noting that the power to issue conclusive certificates under the Commonwealth Act is not confined to the s 36 exemption for internal working documents. Rather, as observed above, the power sits somewhat anomalously alongside similar provisions which deal with documents affecting national security, defence or international relations (s 33(2)), documents affecting relations with the States (s 33A(2)), Cabinet documents (s 34(2)), and Executive Council documents (s 35(2)). A number of the State FOI Acts contain similar provisions, although interestingly none appear to allow for conclusive certificates in relation to internal working documents.[43]
Having said that, the interpretation of s 58(5) accepted by Gleeson CJ, Kirby and Hayne JJ, the result in McKinnon (No 3) notwithstanding, might suggest that there is still some ray of hope for FOI. Notably, these three members of the High Court rejected the majority approach of the Full Federal Court, thus denying that 'reasonable grounds' for the issue of a conclusive certificate would exist if a single reasonable ground, standing alone' could be found. Rather, while avoiding the language of 'balancing' their Honours stressed that the reasonableness of a claim that disclosure of documents would be contrary to the public interest involved a consideration of all relevant matters and arguments. This, while necessarily and correctly falling a long distance short of full merits review, does seem to require a somewhat heightened degree of scrutiny by the AAT of ministerial decisions to issue conclusive certificates.
In this, a majority of the High Court expressly adopted the 'global' approach in outlining the AAT's statutory task under s 58(5). This is a significant development and can be seen to overrule some of the previous authorities on the correct interpretation of s 58(5), notably the Federal Court's decision in Australian Doctors Fund[44] and a number of earlier AAT decisions cited therein.[45]
Nevertheless, it is salutary to recall that the application in McKinnon (No 1) failed. This suggests two things. First, the practical application of the 'global' approach may not often generate a different result from application of the now overruled 'stand alone' approach to the s 58(5) task. Second, and consequently, even with a somewhat heightened level of scrutiny, ministerial decisions to issue conclusive certificates are likely to remain extremely difficult to overturn.[46]
More fundamentally, the exact nature of the review under s 58(5) advocated by Gleeson CJ, Kirby and Hayne JJ remains to be elaborated. It is no simple task to outline a review role for the AAT which goes beyond merely ascertaining that there is at least one reasonable ground for the view that release of documents would be contrary to the public interest and yet refrains from entering into the actual merits of the decision to issue a certificate to that effect. How should a future AAT take into account all relevant matters and arguments, and yet avoid the 'balancing' of those matters and arguments which is the hallmark of the 'merits review' role denied to the Tribunal by s 58(5)?
Here, the analogy put forward by Gleeson CJ and Kirby J requires close attention. As noted above,[47] their Honours used an example of facts which, standing alone, might raise a reasonable suspicion that a person had committed a crime. However, once a 'global' approach is adopted, those facts may lose that character in the face of, for example, a strong alibi. If that alibi is indeed strong, then it negates the formerly reasonable hypothesis of guilt and is simply no longer reasonable to cling to the view that the person in question has committed the crime.
This intuitively demonstrates the superiority of a global approach in this context at least, and appears an excellent example of the 'in principle' superiority of the global approach to the assessment of the reasonableness of a subjectively held opinion, in a wide range of statutory contexts, including the issue of a conclusive certificate. But some caution is necessary. As common lawyers have long understood, the real strength of an argument by analogy depends upon the closeness of that analogy.
However, the analogy adopted by Gleeson CJ and Kirby J is not a particularly compelling one. It is unlikely that there will be many occasions in the specific statutory context in which the AAT is required to undertake its s 58(5) task where an applicant for the release of documents will be able to present the AAT with a comparably compelling 'knock down' refutation of the ministerial assessment of the public interest, a refutation sufficiently strong that the ministerial assessment will similarly be seen as no longer reasonable. The equivalent of the 'cast iron alibi' is likely to be lacking. Rather, the situation is much more likely to be exactly as it was in McKinnon's case where there are arguments in both directions as to the reasonableness of the Minister's opinion, but none of these are sufficiently compelling as to conclusively negate the others. In such a case, s 58(5) would require that the issue of the conclusive certificate be upheld. In the absence of a conclusive refutation of the ministerial opinion, there would remain 'reasonable grounds' supporting the belief that release of documents would be contrary to the public interest, even on the 'global' approach. Thus, even after having taken into account the existence of arguments or grounds weighing against the ministerial opinion, the reasonableness of that opinion would not be negated.
It is that proposition which must be negatived by an applicant, and conclusively so. It would be a rare case where an applicant was able to do this, even applying the 'global' approach to the assessment of reasonableness.
McKinnon (No 3) is undoubtedly of significance in the way in which it develops the notion of what it means for an honestly held subjective opinion or belief to be 'based on reasonable grounds'. It confirms the view previously expressed in George v Rockett that such a belief must be based upon facts and evidence which support it.[48] It goes further than that decision in expressly requiring that an assessment of this reasonable basis must be undertaken in the light of all relevant facts and evidence, not merely those which support the challenged opinion. Thus, it clearly adopts the 'global' approach in place of the now rejected 'stand alone' view. This may be of relevance wherever statutory actions or powers are conditioned upon the existence of a reasonable belief.
For FOI advocates, however, the result is intensely disappointing. Not only did McKinnon lose, but the new test, though seemingly broader, holds little hope that future applications will meet with greater success in the face of a conclusive certificate. Rather, the decisions of both the Federal Court and the High Court clearly points to the substantial weakness of current FOI legislation, particularly as to how it deals with internal working documents. Several observations are in order.
First, the potential breadth of this exemption is considerable. As noted above, s 36 applies to any material 'in the nature of, or relating to, opinion, advice or recommendation obtained, prepared or recorded, or consultation or deliberation that has taken place, in the course of, or for the purposes of, the deliberative processes involved in the functions of an agency or Minister or of the Government of the Commonwealth'. This covers an extremely broad range of material. In particular, McKinnon's case illustrates that it covers precisely those kinds of documents in whose release the public might have a considerable interest, allowing them to better evaluate the policy positions taken by government and the alternative options available to them. This is the kind of information which is the lifeblood of informed democratic discussion. It is not by chance that The Australian newspaper wished to access documents on bracket creep and the First Home Owners Scheme; these were matters of interest to its readers and to the Australian public in general. Yet, on a literal reading of s 36, once any remotely plausible 'public interest' argument against release is put forward to underpin a conclusive certificate they are exempt from release, effectively gutting the Act and robbing it of any real effectiveness in its stated aim of 'creating a general right of access to information in documentary form in the possession of Ministers, departments and public authorities'.[49]
Second, many of the purported 'public interest' arguments against the release of internal working documents are questionable in the extreme. Yet, once solemnly put forward by government counsel, they are routinely accepted with little or no questioning both by the AAT and by courts. Perhaps the greatest disappointment of McKinnon (No 3) is that the High Court so entirely failed to grasp the opportunity presented to it to clarify the nature of the public interest as it applies in the context of FOI legislation and internal working documents in particular.
As noted above, there are two main categories of purported 'public interest' arguments against the release of internal working documents. These are the arguments that internal communication and the efficient working of the public service would be impaired: the 'frankness and candour' arguments, and the claim that the Australian public would be misled by the release of documents, particularly incomplete or superseded documents: the 'misleading the public' arguments. Both categories of argument should be viewed with considerable scepticism.
Three of the reasons given in the conclusive certificates issued in relation to the documents sought by McKinnon fell into the first category. These were question begging in the extreme:[50]
(a) Officers of the Government should be able to communicate directly, freely and confidentially with a responsible Minister and members of the Minister's office on issues which are considered to have ongoing sensitivity and are controversial and which affect the Minister's portfolio.
(b) Officers should be able freely to do in written form what they could otherwise do orally, in circumstances where any oral communication would remain confidential. Such written communications relating to decision-making and policy formulation processes ensure that a proper record is maintained of the considerations taken into account. If they were to be released for public scrutiny, officers may in the future feel reluctant to make a written record, to the detriment of these processes and the public record.
(f) The preparation of possible responses to questions in Parliament is a very sensitive aspect of the work of departmental officers and it is appropriate that briefing and other material produced on a confidential basis in the preparation of those responses, remain undisclosed. The release of such documents would threaten the protection of the Westminster-based system of Government.
These are bald assertions, rather than reasoned arguments. On their face, they are also class claims, making no direct reference to the actual documents at issue, but simply seeking to claim a broad exemption for all documents of which these claims might be made. To the extent that they were supported and particularised in the AAT hearing, this was done by witnesses for the Treasury to whose evidence we are not privy. But a general response may be made.
The 'lack of candour' argument, in its various forms, shows a surprising level of unwillingness to accept the basic principles of accountability. It is an argument which has drawn considerable criticism in the past.[51] Necessarily, accountability may be inconvenient for those who are held to account. This is an 'inconvenience' which is surely inherent in the very concept of accountability. But, with respect, Downes J appeared oblivious to this point when he commented that:[52]
The primary role of government is to govern. Interference with the smooth carrying out of that role will be against the public interest. …there remains a legitimate potential public interest in letting government get on with its role without unnecessary intrusion and distraction.
This is tendentious at the least, begging the question of what is 'unnecessary' intrusion and distraction. It also went beyond any of the arguments advanced on behalf of the Treasurer.
There is also a good deal of empirical experience that suggests that increased accountability positively promotes higher quality administration. A useful analogy may be drawn with the limited duty to give reasons imposed by s 13 of the Commonwealth AD(JR) Act[53] which is now generally recognised as having imposed a positive discipline upon decision makers and improved the overall quality of decision making. Similarly, a recognition that the documents one produces or the advice one gives to a Minister may subsequently be available for public scrutiny is likely to usefully increase, rather than detract from, the incentives for that advice to be of the highest possible quality.
The remaining four 'public interest' reasons given for the issue of the conclusive certificates fell into the 'misleading the public' category. They were:[54]
(c) The release of a document that discusses options that were not settled at the time the document was drafted and that recommends or outlines courses of action that were not ultimately taken has the potential to lead to confusion and to mislead the public. The release of such potentially misleading or confusing material would not make a valuable contribution to the public debate and has the potential to undermine the public integrity of the Government's decision making process by not fairly disclosing reasons for the final position reached. Decision-making processes are multi-layered and the documents reflect partially considered matters and tentative conclusions.
(d) The release of the material would tend to be misleading or confusing in view of its provisional nature, as it may be taken wrongly to represent a final position (which it was not intended to do) and ultimately may not have been used or have been overtaken by subsequent events or further drafts.
(e) The release of documents that contain a different version of estimates, projections, costings and other numerical analysis that cannot be put into context because of the absence of any explanation of the variables used or assumptions relied upon has the potential to lead to confusion and to mislead the public. The release of such potentially misleading or confusing material would not make a valuable contribution to the public debate and has the potential to undermine the public integrity of the Government's decision-making process by not fairly disclosing reasons for the final position reached.
(g) The release of documents that are intended for a specific audience familiar with the technical terms and jargon used, has the potential for public misunderstanding in that the contents of the documents could be misinterpreted. These documents were not intended for publication and publication would be misleading as the documents do not contain sufficient information for an uninformed audience to interpret them correctly and reasonably.
The argument that the Australian public would be misled by the release of internal working documents for these sorts of reasons is not only a profoundly patronising and fundamentally anti-democratic argument — it is certainly both of these things — it also ignores the reality that political discourse in modern states is largely conducted not by individuals working alone but by a wide range of lobby groups, each with special interests in which they invest their resources. These groups are well equipped to understand so called 'misleading' documents and to place them in their appropriate context. It is disappointing indeed that judges, entrusted with the ultimate responsibility for preserving a democratic society against executive excess, should so limply accept these arguments as valid, and without at least requiring a firmer evidential basis for them. A more robust scrutiny of such arguments against release is long overdue. The High Court had the opportunity in McKinnon (No 3) to make an important statement delineating the nature of the public interest in relation to FOI claims, but both majority and minority judgments eschewed doing so.
A great deal of the difficulty lies, of course, in the statutory scheme itself. It is hardly satisfactory that access to information, a fundamental pillar of democratic accountability, should lie so completely in the hands of government Ministers, ie, the very persons whose accountability, amongst others, the Act is designed to facilitate. It is inevitable that Ministers will come to view 'the public interest' through the lens of partisan political self interest, and will issue conclusive certificates accordingly. It is hard to doubt that many of the 'public interest' reasons routinely trotted out by Ministers, of both political persuasions, for exemption from release are more closely related to the potential for political embarrassment than to true notions of the 'public interest' in the context of FOI legislation. The effect of s 58(5) as it currently stands is that such abuse will remain all but unchallengeable.
McKinnon's case demonstrates that the courts remain complicit in this failure. By adopting an unduly literalistic interpretation of the Act, the Australian courts, and the High Court in particular, have failed to act with a full appreciation of their greater constitutional role. Australia is a representative democracy. This is a profound fact. Indeed, it is arguably the single most profound fact as to the nature of our legal order and it is a fact that gives legislation such as the Act, which facilitates the democratic process, a special constitutional significance. It is a fact which justifies a reading and interpretation of that Act which is broad and purposive, and one which protects, enhances and facilitates its democratic function, not one that denudes that function. This requires that the s 36 exemption, and in particular the ability to issue conclusive certificates on 'public interest' grounds, are read narrowly, and that the most fundamental aspect of the 'public interest' referred to in the Act be recognised as the public interest in flourishing democratic accountability and participation.
It appears as though it may be a considerable time before the High Court is willing to adopt such an approach to the Act. Until such time, the outcome of the McKinnon litigation stands as a sad landmark to the weakness of the democratic spirit within Australia, a weakness in which both Executive and Judiciary have played their part.
[*] Senior Lecturer, Law School, University of Adelaide. My thanks are due to the anonymous referee for their comments on an earlier draft. The remaining deficiencies are the sole responsibility of the author.
[1] Australian Capital Television Pty Ltd v Commonwealth (No 2) [1992] HCA 45; (1992) 177 CLR 106, 137 (Mason CJ).
[2] Ibid 231.
[3] Ibid 139.[4] For a discussion of these rationales, see Moira Paterson, Freedom of Information and Privacy in Australia: Government and Information Access in the Modern State (2005) 9-13. The objects of the FOI Act are set out in s 3 of the Act and notably s 3(1)(a) includes 'making available to the public information about the operations of departments and public authorities'. See also Re Cleary [1993] AATA 248; (1993) 18 AAR 83, 87.
[5] See, eg, Australian Law Reform Commission, Open Government: A Review of the Freedom of Information Act 1982, Report No 77 (1995); Ron Fraser, 'Where To Next with the FOI Act? The Need For Renewal — Digging In, Not Giving Up' [2003] AIAdminLawF 15; (2003) 38 AIAL Forum 57.
[6] Australian Law Reform Commission, above n 5, [2.10] - [2.11].
[7] Ibid [6.4] (Recommendation 18).
[8] See, eg, Paterson, above n 4, 303; Rick Snell, 'Conclusive or Ministerial Certificates: An Almost Invisible Blight in FOI Practice' (2004) 109 Freedom of Information Review 9.
[9] It is difficult to get exact statistics. However, in 2003, the Treasurer, Peter Costello, indicated that 55 conclusive certificates had been issued between 1982 and 1986. It appears that records of the issue of these certificates were not kept after 1986: Commonwealth, Parliamentary Debates, House of Representatives, 4 December 2003, 23830–1 (Peter Costello, Treasurer).
[10] Re Howard [1985] AATA 100; (1985) 7 ALD 626.
[11] Re McKinnon [2004] AATA 1364; (2004) 86 ALD 138, 139 [3] ('McKinnon (No 1)').
[12] Ibid.
[13] McKinnon (No 1) [2004] AATA 1364; (2004) 86 ALD 138.
[14] McKinnon v Secretary, Department of Treasury [2005] FCAFC 142; (2005) 145 FCR 70 ('McKinnon (No 2)').
[15] McKinnon v Secretary, Department of Treasury [2006] HCA 45; (2006) 229 ALR 187 ('McKinnon (No 3)').
[16] Ibid 198 [40].
[17] Ibid 188 [1], 192 [17] (Gleeson CJ and Kirby J), 202 [54] (Hayne J).
[18] McKinnon (No 2) [2005] FCAFC 142; (2005) 145 FCR 70, 74 [5] (Tamberlin J), 129 [160] (Jacobson J) and McKinnon (No 1) [2004] AATA 1364; (2004) 86 ALD 138, 141 [14] (Downes J).
[19] McKinnon (No 2) [2005] FCAFC 142; (2005) 145 FCR 70, 139 [232] (Jacobson J).
[20] Ibid 140 [233].
[21] Australian Doctors' Fund Limited v Commonwealth of Australia [1994] FCA 1053; (1994) 49 FCR 478.
[22] Attorney-General's Department v Cockcroft [1986] FCA 35; (1986) 10 FCR 180, 190 (Bowen CJ and Beaumont J). [23] McKinnon (No 2) [2005] FCAFC 142; (2005) 145 FCR 70, 74 [5]-[6] (emphasis in original). [24] Ibid 76-7 [16].[25] George v Rockett [1990] HCA 26; (1990) 170 CLR 104.
[26] Ibid 112.
[27] McKinnon (No 2) [2005] FCAFC 142; (2005) 145 FCR 70, 92 [50] (emphasis added).
[28] Ibid 89 [45].
[29] McKinnon (No 3) [2006] HCA 45; (2006) 229 ALR 187, 190 [11].
[30] Ibid 191 [12].
[31] Ibid 221 [129].
[32] Ibid 222 [131].
[33] Ibid 202-3 [56].
[34] Ibid 204 [63] (Hayne J).
[35] McKinnon (No 1) [2004] AATA 1364; (2004) 86 ALD 138, 150 [54].
[36] Ibid 150 [56].
[37] Ibid 152 [66].
[38] Ibid 136 [34]. Observing that 'although claims based on frankness and candour have not been well received the same cannot be said for claims based on the need for confidentiality.'
[39] Ibid 150 [52].
[40] Ibid 152 [66].
[41] See, eg, ibid 155 [77].
[42] The failure of government to respond to calls for FOI reform is well documented. One such call, which recommended the removal of conclusive certificates in relation to internal working documents, was made by the Australian Law Reform Commission: above n 5, [9.19] (Recommendation 53A).
[43] Under the New South Wales Freedom of Information Act 1989 (NSW), 'Ministerial certificates' can be issued in relation to Cabinet documents, Executive Council documents, documents affecting law enforcement and public safety, and documents affecting counter terrorism measures and 'shall … be taken to be conclusive evidence that the document is a restricted document': s 59(1). There are no conclusive certificates in relation to internal working documents. Similarly, the Victorian Freedom of Information Act 1982 (Vic) allows the issue of certificates (not expressly stated to be conclusive) that 'establishes' that a document is an 'exempt' Cabinet document: s 28(4). A similar provision in s 29A(2) applies to documents affecting national security, defence or international relations. Again, there is no provision for conclusive certificates in relation to internal working documents. Nor do the other State jurisdictions appear to provide for such certificates in relation to internal working documents.
[44] Australian Doctors' Fund Limited v Commonwealth of Australia [1994] FCA 1053; (1994) 49 FCR 478.
[45] See, eg, Re Porter (1988) 8 AAR 335; Re Waterford (No 2) [1984] AATA 67; (1984) 1 AAR 1; Re Rae (1986) 12 ALD 589, cited in ibid 484-5.
[46] The effect of s 58A of the Act should also be noted, whereby the AAT's powers in reviewing a conclusive certificate under s 58(5) are recommendatory only.
[47] McKinnon (No 3) [2006] HCA 45; (2006) 229 ALR 187, 191 [12].
[48] George v Rockett [1990] HCA 26; (1990) 170 CLR 104, 112.
[49] Freedom of Information Act 1982 (Cth) s 3(1)(b).
[50] These were reproduced at McKinnon (No 1) [2004] AATA 1364; (2004) 86 ALD 138, 114 [26].
[51] See, eg, Rick Snell, 'The Ballad of Frank and Candour: Trying to Shake the Secrecy Blues from the Heart of Government' (1995) 57 Freedom of Information Review 34; Peter Bayne and Kim Rubinstein, 'Freedom of Information and Democracy: A Return to the Basics?' (1994) 1(2) Australian Journal of Administrative Law 107.
[52] McKinnon (No 1) [2004] AATA 1364; (2004) 86 ALD 138, 151 [59]-[60].
[53] Administrative Decisions (Judicial Review) Act 1977 (Cth).
[54] McKinnon (No 1) [2004] AATA 1364; (2004) 86 ALD 138, 114 [26].
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/FedLawRw/2007/11.html