AustLII Home | Databases | WorldLII | Search | Feedback

Alternative Law Journal

Alternative Law Journals (AltLJ)
You are here:  AustLII >> Databases >> Alternative Law Journal >> 2001 >> [2001] AltLawJl 71

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

De Maria, William --- "FoI unshackled? The Murray Bill" [2001] AltLawJl 71; (2001) 26(4) Alternative Law Journal 191

FoI UNSHACKLED?: The Murray Bill

William De Maria

A detailed analysis of proposed freedom of information reforms.

…unless citizens have the power to access and independently scrutinise government information there is little prospect of having a genuinely deliberative and participatory democracy.

Senator Andrew Murray

FoI in chains

Secrecy is the high-octane fuel that power runs on. It has always been so. One of the many negative inheritances from our British history has been the obsession with official secrecy.[1] Over time this fixation has poisoned the wells of democracy, and remains to this day a worrying fixture of public and private sector life in Australia.[2]

Australia’s freedom of information Acts were intended as a response to the secretive way governments did business in the 1960s and 1970s.[3] Now we have government doing very different business. A government of basic service-deliverers is mutating into a government of traders and regulators. The FoI Acts have not kept up with this development.[4] Previous community demand for openness has, to some extent, given over to concerns about the human costs of globalisation, de-regulation, national competition policy and outsourcing (big voting issues in the Western Australian and Queensland elections of 2001). This has served to distract the reform process away from issues of freedom of information.

The result has been that the nation’s FoI Acts are languishing. Commonwealth statistics show that while 565,219 access requests have been processed between the date of commencement of the Commonwealth Freedom of Information Act on 1 December 1982 and 30 June 2000, there has been no growth in applications for the last four years. In fact there were 1700 fewer access requests in 1999/2000 than the previous year.[5] This stagnation may be attributed to a number of things: alternative sources of information, proactive release policies,[6] and improved public Internet access.

It may also imply that there are structural and cultural obstacles to access in the Act and in the behaviour of public servants who manage access decision making. The 20-year process of bureaucratic co-option of FoI is now complete. FoI, once a revolutionary idea, has succumbed to ordinariness. FoI administration is often in the hands of come-and-go bureaucrats who are not personally fired-up by the concepts of official openness and transparency. This converts into cautious and drawn out applicant–FoI administrator transactions, and review and appeal mechanisms that drag the huge ball and chain of legal formalism.[7]

As disturbing, are the differential access policies between ‘personal’ and ‘policy’ applications. With respect to personal information applications (in excess of 87% of all Commonwealth requests in 1999/2000)[8] the FoI Acts work reasonably well. But that is a bit like applauding a lift because it takes you up and down; that is what it is supposed to do. The true test is whether FoI allows citizens access to contemporary governance; to enter, in other words, the world of official policy.[9] On that test the FoI Acts at best get mixed results.

The statistics in this area can be quite deceptive. In 1999/2000, 4.79% of all access requests under the Commonwealth Act were refused outright. This figure is kept small by the huge number of grants in full to personal information claims.[10] What needs to be drawn out from the low rejection statistic is the high refusal rates on an agency basis. Some examples. In 1999/2000 Telstra rejected 32% of applications (148); Department of Employment, Workplace Relations and Small Business rejected 30% of applications (73); and Australian Customs Service rejected 30% of applications (48).[11]

An initiative currently before the Commonwealth Parliament may override some of the criticisms mentioned above. I refer to Senator Andrew Murray’s Freedom of Information Amendment (Open Government) Bill 2000.[12] Senator Murray’s Bill is in lieu of the Howard Government’s failure to implement the recommendations of the 1996 review of the Act carried out by the Australian Law Reform Commission and the Administrative Review Council.[13] Senator Murray’s Bill has been referred to as the most significant reform proposal since the inception of the Commonwealth FoI Act in 1982.[14] This article critically examines his proposal. While it is a mixed review of robust applause and slow hand clapping, Senator Murray is to be congratulated for taking in the orphaned Act and attempting its transformation into a true plinth stone of democracy.

First the good news …

There is something very important (but not entirely unique) about the way Senator Murray starts his FoI reform Bill. The objects section of a statute is like the doorway to a house. The objects section of the current Act guides people through a narrow portal to a meanly furnished array of access rights. Proceeding further one finds a room packed to the ceiling with exemptions. The Act talks of what information citizens can have: ‘public information about the operations of departments …’ Senator Murray’s substitution talks of why we need it:

3(1)
(a) to enable participation in the policy, accountability and decision-making processes of government;
(b) to open the government’s activities to scrutiny, discussion, comment and review; and
(c) to increase the accountability of the executive branch of government

These are followed by a restatement of rights already conferred in the Act; rights to personal information and to bring about amendments to personal information held in documentary form. Of interest is a new object, to create ‘a general right of access to the national resource of information in documentary form in the possession of government’.

‘Participation’, ‘openness’, ‘accountability’, ‘the national resource of information’ — what powerful and relevant concepts to open up the Commonwealth FoI Act! While these vision markers lack interpretative power they can have enormous symbolic force.[15] So too can any provision that addresses the unspoken, the informal administration of the Act. Senator Murray adds to the Bill (in the interpretation section, oddly enough) a legal requirement to provide information, even though it is embarrassing to government. Senior bureaucrats are the ones who normally do the FoI policy setting in agencies and conduct internal reviews. It goes without saying that having successfully embraced the agency ethos they will be disinclined to execute a lawful process of discovery under FoI if such would expose the agency to attack, ridicule or censure. While the Murray provision is not strictly necessary, as no exemption on this ground exists, the provision serves as a timely moral reminder of the obligations of FoI officials to do the right thing.[16]

A number of provisions in the Murray Bill reduce the statutory time limits on processing requests, reviews and external appeals. This can only be a good thing. The Commonwealth public sector and government-owned corporations (GOCs) have had almost 20 years to develop efficient procedures for the identification of agency information, assessment of access status and review and appeals on those assessments. Although information technology has enhanced efficiency of access to information, the current situation is that only 76.77% of FoI requests were processed in the 30-day period mandated by s.15(5)(b).[17] This period can be extended under certain exigencies to 60 days.[18] Even then, 7.6% of all requests (that is, 2415 out of a total 31,784 requests in 1999/2000) took more than 60 days to respond to.[19]

Senator Murray tips his lance at one of the great fortifications in the Act — s.37: documents affecting enforcement of law and protection of public safety. This section provides sensible barriers to the release of information if it would disclose ongoing criminal investigations, break the cover of informants or compromise trials. The Murray Bill proposes to add to s.37(2):

or, (d) prejudice the security of a lawful place of detention

I have two problems with this proposal. First is it necessary? Section 37(1) of the Act treats a document as exempt if its disclosure ‘would prejudice the enforcement or proper administration of the law …’ Doesn’t this capture the security of lawful places of detention?

Second, with the administration of more and more places of detention being out-sourced (the troubled Woomera Refugee Detention Centre controversially run for the Commonwealth Department of Immigration and Multicultural Affairs by Australasian Correctional Management comes to mind) do we need another reason to allow these organisations to escape public scrutiny? Embedded in this observation is the failure of FoI to reach into the administration of private companies on contract to the government. This issue is considered below.

The law enforcement and public safety exemption of the FoI Act has been virtually untouchable. There is an emotional barrier there; a fear that civil disorder will occur if these exemptions are diluted. Senator Murray makes an important advance on this exemption by proposing that the following classes of documents be de-classified:

• those pertaining to illegal police operations;

• those that describe in general terms the structure of law enforcement programs;

• those that provide evaluations of law enforcement programs; and

• those that are routine law enforcement reports.

The Murray Bill identifies one of the great contradictions in the FoI Act: the presence of a powerful exemption in respect of secrecy provisions. The proposed repeal of this entire section is one of the most important innovations in the Bill. The fixation with secrecy, as I mentioned at the outset, is still very much a part of public life in Australia.[20]There are over 150 secrecy provisions in Commonwealth Acts and regulations,[21] over 100 such provisions in Western Australian law[22] and 160 secrecy provisions in Queensland law.[23]

Finally, the addition of s.58(2)(A) is welcome. This new subsection allows the Administrative Appeals Tribunal (AAT) to grant access to confidential business documents if they can establish a public interest override. One could also argue that if this power is legitimate, then why not give the AAT a general public interest over-ride on all exempt material?

And now the bad news …

I would say that the biggest deficiency in Senator Murray’s Bill is that it strikes at the insidious exemptions part of the Act with a soft pillow. To be fair, this is an academic talking. Senator Murray’s efforts may be right on the edge of political feasibility. Certainly in parliamentary circles he and his small group must continually face off the shocking indifference displayed by the major parties to FoI reform.

It is well known that FoI administrators tend to hedge their bets by citing as many exemptions as possible. It can be quite intimidating for an FoI applicant to receive a rejection citing numerous exempt provisions in the Act, with each exemption accompanied by long pro forma insertions of the alleged relevant case law.

The question is, do we need the exemptions? One option, albeit radical, is to delete all exemptions within the Act and replace them with a single harm test. This would mean that if the release of a requested document would, in the agency’s view, cause social or economic harm (closely defined), then the application is simply refused. The agency does not need to attempt to justify how refusal is consistent with an existing suite of exemptions. What it must do, however, is precisely state:

• what social or economic harm would the release of the document cause?

• how real is the possibility of harm?

• to whom or to what, would the envisaged harm occur?

• which factors it took into account in determining the above.

Applicants would then have the same choice as they have now, accept the ruling or appeal it.

Senator Murray does not go that far, and I am not critical of his Bill for that reason. However, his proposed reforms of the exemptions often dance around the main issues. For instance he would repeal the existing section dealing with the exemption of documents of a sensitive interstate nature and replace it with a section that abolishes the power of ministers to issue conclusive (non-appealable) certificates regarding State-to-State documents. This is a good thing. The ministerial certificate is just a pleasant re-invention of ‘iron rule’. It has no place in an FoI Act. This innovation is counter-pointed with the re-inclusion of the exemption of sensitive interstate documents. This exemption is sorely in need of challenge. It protects an enormous amount of clandestine government-to-government dialogue that should be released in the public domain.

The Bill adopts the same reform approach to s.34 (Cabinet documents). Admittedly taking on this section is not for the faint hearted. The Bill’s repeal of s.34(1)(a), and its replacement with a succinct:

A document is an exempt document if it is:
(a) a document that was brought into existence for the purpose of submission to Cabinet for its consideration

is a very important, but ultimately cautious step. The exploitation of the Cabinet exemption provision to block legitimate access to public documents is part of the sorry history of FoI in Australia, more so in the jurisdictions of Victoria and Queensland.

Further, the proposed intention of ss.34(1)(AA), to clip ten years off the traditional embargo time of 30 years for the legal release of Cabinet documents, does not go far enough. Is there a plausible public interest justification for quarantining Cabinet documents at all?[24] An exception-to- the-rule argument can be made that documents directly pertaining to economic, diplomatic and national security decisions of Cabinet should be kept out of the public domain. But then only as long as the threat to the national interest prevails (as determined by an external reviewer such as the AAT). But that argument would apply to only a very small class of documents. What we have now is a long quarantine period for all Cabinet documents. The only justification for this broad approach is a fear that disclosure would expose government to opposition, media and interest group scrutiny and attack within the context of the adversarial party system. Heh, haven’t I just described the democratic process?

What about pre-Cabinet? FoI statistics in all jurisdictions show the deliberative process to be one of the most common reasons for refusing access. The Bill leaves this unwarrantedly powerful exemption intact and only repeals sections to do with ministerial certificates. While under s.36(1)(b) of the Act a public interest case must be made, experience suggests that the case is usually a set of clichés about the importance of keeping the advice-proffering process ‘fearless and frank’. The enormous changes in the public service since the Act’s inception in 1982, when considered in the framework of this point, suggest that much of the deliberative process is meeting political rather than policy designs. Further much advice is now outsourced to ‘experts’ beyond the normal accountability loop. This has opened the door to a new and unanticipated reason to withhold; that the expert advice remains the intellectual property of the consultant, and as private information it is quarantined from FoI discovery.

It is time for the deliberative process exemption to be repealed altogether. The division between information about government programs and policies and information about how these policies and programs were formed (or abandoned as the case may be) is spurious and indefensible.[25]

What of business secrets? No reform of an Australian FoI Act can be taken seriously these days unless it curtails the ever-widening ambit of business exemptions. The Murray Bill envisages some changes; releasing documents relating to research and documents relating to the national economy from the scope of this exemption. However the Bill (despite the addition of ss. 58(2)(A), see above), is an inadequate response to the commercial-in-confidence (CIC) ‘plague’ we are currently facing.

There is no doubt that CIC is used excessively.[26] In 1997 the CIC exemption was used by FoI administrators in Queensland 21,242 times to deny access to documents held by government.[27] We do not know how 11 people died in privatised prisons in Victoria in 1997 because that information, requested under FoI, was treated as CIC. The water privatisation contract in South Australia runs for 20 years and we do not know what is in it. Why? CIC again. One of the world’s largest outsourcing contracts, the Commonwealth Government’s Job Network, is terra incognita to us because of CIC.

A very important motion by Senator Murray on government contract disclosure is working its way through the Senate. It is a shame he did not tackle the problem in his FoI Bill. There is so much we could do here, and there is support: auditors-general, parliamentary committees, ombudsmen and information commissioners.[28] All saying the same thing, all calling for reform. While CIC may be good for business, it is inimical to the fragile workings of democracy. I think the message should be ‘if you wish to do business with the people there can be no confidentiality and no secrets’.

Another issue not picked up in the Murray Bill is the ‘sufficiency of search’ concept. The question here is how do we know whether FoI officials have made thorough searches of their holdings for the relevant documents, particularly if those documents contain sensitive political matter? The answer is we don’t, unless we know of the existence of a document prior to an application for access to it. In most situations we are in the hands of the FoI officials.

What assurances exist in the Commonwealth Act that document searches are sufficient if not exhaustive? Section 24(A) of the Commonwealth FoI Act states that an agency or a Minister must take ‘reasonable steps’ to find the document. In what may have been an oversight, this requirement is deleted from the Murray Bill. The Senate Committee has accepted his recommendation. It seems that the only way to ensure sufficiency of search is for agencies to publish lists of all files in its possession along with annotated indexes of folios within these files.

Part VI of the Bill responds to the vexed question of the supervision of FoI legislation by providing for an Information Commissioner. Unfortunately, Part VI depends on a frail administrative model, when what is needed is a forceful, bi-partisan-supported agency that will rescue FoI, and fight for access rights.

Examples of the inherent weakness in the set up of the proposed Information Commissioner are not difficult to find. It appears that the only determinative power the proposed Commissioner has is in the setting of charges (s.66(J)(f)). A saving grace is the requirement that FoI decision makers and the AAT must take into account the Commissioner’s guidelines (s.66L).

What does it mean to provide the proposed Information Commissioner with legislative firepower? First is it necessary? I would say it is. The CIC avoidance strategy discussed above provides the clearest indication of the undermining of the processes of openness and transparency. But there are many more. The aforementioned Commonwealth Ombudsman’s report into the Act and the recently completed investigation by the Commonwealth Auditor-General into the usage of CIC bespeaks of a good deal of official unease about the Act’s improper administration.

An FoI Commissioner requires a stronger legislative base that would:

allow investigations of breaches of the objectives of the Act by agencies. The Commissioner would be able to adversely mention these agencies in reports, and in serious cases, initiate a brief for the charging of agencies in courts of law.

enable the Commissioner to establish and monitor a stronger proactive release policy. There is something profoundly undemocratic about citizens having to ask for official information, more so when the request involves drawn out, formal and complicated processes. At present, agencies can release information outside the Act. It is a discretionary power, rarely used with respect to sensitive policy material. We need greater outside-the-Act mandatory release of official information. The compulsory release provisions in the British Colombia Freedom of Information and Protection of Privacy Act with respect to material that informs the public of significant environmental and safety issues comes to mind. The news that a toxic leak polluted parts of Kakadu National Park, came to us earlier this year, one month after it happened. This is the sort of disclosure that would have been made instantly had mandatory public reporting provisions been in place.

A web and e-mail based mandatory disclosure program could radically reduce the scope of the FoI Act. The Act would then be the statutory gate through which contentious material would be released if it passed the public interest test.

Current developments

The Senate Legal and Constitutional Legislation Committee’s deliberation on the Murray Bill was a mixed result for FoI reform. During its considerations Senator Murray withdrew a number of important innovations with respect to exemptions, including:

• a public interest test on access to documents affecting relations between the States;

• a ten-year reduction in the Cabinet quarantine period;

• the repeal of ministerial certificates for internal working documents;

• the reform of the law enforcement exemption;

• the deletion of the secrecy enactment exemption;

• the stipulation that embarrassment to government is an irrelevant criterion for exemption;

• the proposed power of the AAT to apply a public interest test to business exemption appeals.

As a result of the abandonment of the above exemptions and the ineffective proposal for an Information Commissioner, the Murray Bill is seriously deficient. The second reading debate on the Freedom of Information (Open Government) Amendment Bill 2000 is due to resume in the latter part of 2001. There is a real chance that this time frame could be disturbed by the forthcoming federal election. Unless a major party supports the Murray initiative, it is on its way to the graveyard of private member’s Bills. Flawed and cautious as it is, this well-meaning Bill is, at least, an attack on government secrecy. The struggle for a truly effective Commonwealth FoI Act continues.


[1] The United Kingdom was one of the last Western countries to implement freedom of information legislation.

[2] See Terrill, G., Secrecy and Openness: The Federal Government from Menzies to Whitlam and Beyond, Melbourne University Press, 2000.

[3] It is sobering to note that the platform of ‘open government’ was used only for the first time in Australia in the 1972 federal election campaign.

[4] See De Maria, W., Submission to Queensland Legislative Assembly, Constitutional Law and Administrative Review Committee, Review of the Freedom of Information Act, 12 May 2000; De Maria, W., ‘Commercial-in-Confidence: Obituary to Transparency’, Keynote Address to CPA (Queensland Branch) Audit Intensive Day, 10 November 2000.

[5] Commonwealth of Australia, Attorney-General’s Department, Freedom of Information Act 1982, Annual Report 1999–2000, p.2.

[6] Under ss.8 and 9 of the Commonwealth FoI Act.

[7] See Commonwealth Ombudsman’s evidence to Senate Legal and Constitutional Legislation Committee, Inquiry into the Freedom of Information Amendment (Open Government) Bill 2000, April 2001.

[8] It is still impossible to distinguish between ‘personal’ and ‘policy’ access requests in the Commonwealth FoI statistics. We do know that 87% of all access requests in 1999/2000 were made to just three agencies: Department of Veterans Affairs, Department of Immigration and Multicultural Affairs, and Centrelink. On the whole these were non-contentious requests for personal information. See Commonwealth of Australia, Attorney-General’s Department, Freedom of Information Act 1982, Annual Report 1999–2000, p.6.

[9] The Commonwealth Ombudsman made this point in her own motion inquiry into the operation of the FoI Act. See Commonwealth Ombudsman, Need to Know, 3 June 1999. The Attorney-General has announced that the government will accept recommendation 11 from this report. In future, agencies will have to distinguish between personal and non-personal requests in their statistical accounting.

[10] For example the Department of Veteran Affairs had a grant in full rate of 99.57% in 1999/2000. See Attorney-Generals Department, Freedom of Information Annual Report, 1999–2000, p.6.

[11] Attorney-Generals Department, above, ref 10.

[12] The Bill was introduced into the Senate on 5 September 2000 as a private members Bill. It was referred to the Legal and Constitutional Law Committee on 11 October 2000. That Committee was scheduled to report by 31 March 2001. The Committee reported in April 2001. Senator Andrew Murray is an Australian Democrats representative from Western Australia.

[13] According to the ALRC, the government has not yet responded to the Commission’s 1996 FoI report. See ALRC evidence to Senate Legal and Constitutional Legislation Committee, Inquiry into the Freedom of Information Amendment (Open Government) Bill 2000, April 2001, p.18, fn 16.

[14] Senate Legal and Constitutional Legislation Committee, Inquiry into the Freedom of Information Amendment (Open Government) Bill 2000, April 2001, p.1.

[15] The courts are in conflict regarding this point. Some decisions do not interpret the object sections of Acts as predisposing towards disclosure. See News Corporation Ltd v National Companies and Securities Commission (1984) 1 FCR 64 and Searle Australia v Public Interest Advocacy Centre [1992] FCA 241; (1992) 36 FCR 111 (which both rejected such a purposive approach). Cf Kirby J’s approach in Commissioner of Police v District Court of NSW (1993) 31 NSWLR 606. The author would like to thank the referee for drawing attention to these cases.

[16] FoI case law has no enthusiasm for factoring embarrassment to government into the determination of the public interest. In its supplementary submission to the Senate Legal and Constitutional Legislation Committee, Inquiry into the Freedom of Information Amendment (Open Government) Bill 2000, the Attorney-General’s Department mentioned a number of leading cases. See for example, Harris v ABC [1983] FCA 242; (1983) 50 ALR 551 and Marr v Telstra Corporation (1993) FoIR 70. However see Re Howard & Treasurer of the Commonwealth [1985] AATA 100; (1985) 7 ALD 626 where the Court held, inter alia, that in balancing factors favouring disclosure against those favouring non-disclosure it was relevant to take into account the fact that the ‘frankness and candour’ of public servants might be inhibited (by embarrassment or otherwise) if their internal deliberations were disclosed. The author wishes to thank the referee for drawing attention to this case.

[17] Attorney-General’s Department, Freedom of Information Annual Report, 1999–2000, p.9.

[18] For example s.26A requires consultations with States on release of certain documents.

[19] Attorney-Generals Department, Freedom of Information Annual Report, 1999–2000, p.9.

[20] See Terrill, G., above. See also Finn, P., Official Information, Integrity in Government Project, Interim Report 1, 1991.

[21] See McGinness, J., ‘Secrecy Provisions in Commonwealth Legislation’[1990] FedLawRw 3; , (1990) 19 Federal Law Review 49.

[22] Western Australian Commission on Government, Report 1, Chapter 2, point 2.3.1.1.

[23] Queensland Law Reform Commission, The Freedom of Information Act 1992. Review of Secrecy Provision Exemption, Report No. 46, March 1994, Appendix C.

[24] On 26 April 2000 the Welsh Cabinet minutes were published, and posted on the Internet for the first time, six weeks after the Cabinet meeting took place.

[25] The same point was made recently by the UK-based public interest lobby group Campaign for Freedom of Information, with respect to the UK FoI Act.

[26] There is evidence that the CIC problem has been in public administration for a long time, certainly since the 1970s. See Senate Finance and Public Administration References Committee, Inquiry into the Mechanism for Providing Accountability to the Senate in Relation to Government Contract, p.1. However the current outsourcing policy puts a new urgency on the matter.

[27] De Maria, W., ‘Revealing State Secrets’, Courier-Mail, 12 April 1999, p.13; De Maria, W., ‘Democracy in Eclipse’, Courier-Mail, 5 January 2000, p.11. Statistics from Queensland Department of Justice, Freedom of Information Act, Annual Report, 1996–97, Appendix F.

[28] For a review of the mounting pressure against CIC see De Maria, W., ‘Commercial-in-Confidence: Obituary to Transparency’?, Australian Journal of Public Administration, in press.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/AltLawJl/2001/71.html