Alternative Law Journal
When it was passed 25 years ago the Commonwealth Freedom of Information Act 1982 broke significant new ground in seeking to overturn a deeply entrenched tradition of government secrecy. Although freedom of information (Fol) had origins in 18th century Sweden and had been established in the United States in 1966, the Australian Act was the first national legislation of its kind to be introduced in a country with a Westminster-style system of government. In fact its enactment was preceded by considerable concern as to whether it could be implemented without causing harm to its very different constitutional and administrative frameworks. Today those doubts have long been put to rest. Australian society and politics have moved a long way since those early tentative years and Fol legislation now forms an integral part of the democratic framework throughout Australia as well as in many other countries with Westminster-style systems of government, including even the UK. That is not to say, however, that the Act has lived up to its promise of delivering open government.
A major difficulty in attempting to assess its overall impact is the lack of useful empirical data concerning its operation. Although the Attorney-General's Department collects annual statistics, these do not provide sufficient detail to allow for any extensive analysis and are generally incomplete. The Act was, however, extensively reviewed in 1994-5 in a joint report by the Australian Law Reform Commission and Administrative Review Council, Open Government: A Review of the Federal Freedom of Information Act ('the Open Government report').That report contained a large number of recommendations for reform, many of which were supported in an 'own motion' report prepared by the Commonwealth Ombudsman in 1999 but none have yet been implemented.
In 200 I Senator Murray introduced the Freedom of Information (Open Government) Bill 200 I which sought to give effect to the more significant of these recommendations. That Bill was reviewed in a report by the Senate Legal and Constitutional Committee which recommended that the Bill should proceed subject to a number of changes. More recently, Senator Murray has introduced another more modest version of that Bill (the Freedom of Information (Open Government) Bill 2003 (Cth) ('the 2003 Bill') which has been edited to incorporate those recommended changes.
This article draws extensively from each of these resources as well as from reviews of freedom of information legislation in other countries. It argues that, although the Bill is an important step forward, it fails more fundamentally to overcome the 'request and wait' model by using today's technology to push out information to the public. It also argues that our Fol ideas need now to be reconceptualised in terms of power rather than continue to be seen in terms of the old traditional democratic notion applied to a smaller public sector.
The Commonwealth Act, like its state and territory counterparts, has three different features. First, it requires the publication of specified information with a view to promoting greater knowledge of the affairs of government agencies, the procedures for exercising Fol rights and rights to participate in decision making. Second, it provides for a universal right of access to documents in the possession of government agencies as a means for ensuring greater political accountability. Third, it also provides a mechanism for the exercise of two important information privacy rights: the right to access one's own personal information and the right to require its amendment if it is not correct.
If one takes the levels of usage of the Act as a crude measure of its performance it seems to be reasonably successful as a device for facilitating the exercise of information privacy rights but not in promoting its other objectives. For example, statistics for the year ending 30 June 2001 show that requests for personal information made up approximately 90% of all requests. It is not clear what proportion of the I0% of requests for other information were made by the media or public interest groups but evidence suggests that media usage of Follaws in Australia is very low. This contrasts with the position in Ireland, for example, where requests for personal information make up approximately 45% of requests and requests by the media currently make up a substantial 20%.
Interestingly, also, it seems that use of the Act by businesses has been comparatively low as compared with the United States and Canada For example, in the case of Canada statistics for 200 I show that business users accounted for nearly 41% of all requests for access and that many people sought access to information to assist them to win government contracts. Likewise, it is generally accepted that businesses are the largest category of users in the United States.
It is an important aspect of the Act that agencies are expected to do more than simply accede to requests for access to information. They are required also to actively publish and make available specified categories of information.
There are two different dimensions to the publication requirements in the Act. The first is the obligation to publish information which facilitates the exercise of access and amendment rights and participation in the processes of government (s B).This obligation is designed not only to enhance the goals of public accountability and information privacy inherent in the access and amendment provisions, but also to promote public participation thereby giving individuals a greater say in public sector policy making. A major shortcoming, however, is that the Act simply requires that these statements are published in the annual report of the agency or of the portfolio department to which the agency relates. It neither provides for any centralised coordination of their publication nor contains any mechanism for enforcing compliance.Consequently, this aspect of its operation has been one of the most disappointing. The information is not always made available as required, its existence is not well known and, not surprisingly therefore, it is not well utilised.  Moreover; there has been an overwhelming failure to make effective use of new technologies to ensure the widespread dissemination of this information.
The second dimension is the requirement to make available the internal guidelines used by agencies to assist them in the exercise of statutory powers that impact on individuals. Agencies are required to make available at National Archives of Australia offices in all capital cities a statement listing documents used by the agency in making decisions that affect the public. They are also required to make available for inspection and purchase all documents listed in the section 9 statements. The publication of so-called 'internal laws' serves to further the objectives inherent in the 'new administrative law' package to which the Commonwealth Act belongs. They allow individuals to understand the rules which guide the making of decisions that affect them and to ensure that these are both adequate and legally valid. Consequently they are extensively used by administrative lawyers. Although again there is no specific enforcement mechanism, s I 0( I) provides for a limited reversal of the rule against estoppel, which generally applies to administrative actions, where there has been a failure to publish an internal law as required.
The provision of a right of universal access gives effect to what is arguably the core Fol objective -to make the activities of governments more transparent and thereby enhance the operation of our democratic system of government and make the government more directly accountable to the people lf this is to achieved, then it is important that the Act is used extensively as a means for shedding light on activities relevant to political accountability, especially by journalists. That this has not been the case would appear to be due to the fact that it is not structured in a way which makes it attractive to use, that it contains too many inherent restrictions on the information which it makes available, and that it is not well administered.
Two key factors which have affected the Act's attractiveness as a source of information for journalists and others wishing to unearth and to publicise information of general public interest are the length of time that it takes to acquire information and the costs of access. The fact that agencies may take up to 30 days (s 15(5)) (or up to 60 days in the case of requests requiring consultation with third parties (s 15(6)), and that these time limits are not always adhered to, means that issues may no longer be of current interest by the time information about them becomes available. This contrasts with the position under the US Freedom of Information Act which provides for expedited searches where the request is made by a person primarily engaged in disseminating information to the public and the information sought is urgently need to inform the public about some actual or alleged federal government activity.
It is also clear that the costs of access, which were increased significantly in 1986 have acted as a deterrent to its use. Applicants are required to pay $15 per hour for the time involved in searching for documents and $20 per hour for the time spent in deciding whether or not to grant access, including time for examining documents and consulting with third parties. These charges are additional to application fees and photocopying charges and there are no automatic reductions or waivers for journalists or public interest groups. The current fee structure is consistent with a user-pays philosophy but it provides little incentive for agencies to address inefficiencies in records management and produces charges which may be both very substantial and difficult to predict. Again this situation compares unfavourably with that in the United States where charges for search and retrieval do not apply to requests for access by educational or non-commercial scientific institutions whose purposes are scholarly or scientific research, or by representatives of the news media unless they are made for commercial purposes.
Arguably the most significant defect in the Act is the fact that its drafting fails to give effect to the principle of maximum disclosure. In particular, its objects clause is unnecessarily ambiguous and has left it open to the courts to reject an approach to interpretation which leans in favour of disclosure. This contrasts with the position in New Zealand where the objects clause is supplemented by a principle of availability which establishes the rule that information should be made available unless there is a good reason for withholding it.
In addition, although the Commonwealth Act makes it clear that decision-makers retain the discretion to grant access to exempt documents (s 11), the provisions which protect agencies against civil and criminal liability arising from disclosures (ss 91 and 92) do not extend to the disclosure of documents which are believed to be exempt or otherwise not positively required to be disclosed. Furthermore, the combination of exemption provisions which lack a separate public interest test, the ability of ministers to issue conclusive certificates, and an external review body which lacks the ability to grant access to an exempt document or to overturn conclusive certificates means that documents may be withheld from access even where there are strong public interest grounds favouring disclosure. The decision to grant a conclusive certificate is subject to judicial review under the Administrative Decisions Judicial Review) Act 1977 but may be challenged only on the basis of deficiencies in the manner in which the decision was arrived at (as opposed to the matters which the certificate asserts).
Insofar as the drafting of the exemption provisions is concerned, the principle of maximum disclosure suggests that documents should be exempt from access only where it can be demonstrated that their disclosure will result in substantial harm which is not outweighed by a countervailing public interest in disclosure. However, the Act still contains provisions which provide for exemption solely because a document falls within one of the classes described (eg ss 34 and 35). Moreover, it contains a significant number of harm-based provisions which require neither the demonstration of substantial detriment nor the evaluation of the competing interests for and against disclosure. For example, an agency can withhold a document on the ground that it could reasonably be expected to cause some damage to the security of the Commonwealth, irrespective of the fact there may be very strong public interest grounds for requiring the granting of access to it. It can also support such a claim with a conclusive certificate which is not subject to ordinary merits review by the
Commonwealth MT. Although this has not proved to be especially controversial up until now, this state of affairs is likely to change given the renewed emphasis on national security in the aftermath of the Bali bombings.
The absence of separate public interest tests is also problematic in the case of exemptions such as those in ss 42 and 45 which are based on common law tests. The judicial abandonment of the sole purpose test has had the indirect consequence of broadening the legal professional privilege exemption which may be used to withhold information provided by in-house lawyers even where there is a clear public interest in its public disclosure. In the case of s 45 it is open to argument that public interest is a relevant factor in determining whether or not there has been an actionable breach of confidence, at least in those cases where the information in question relates to some iniquity or a threat to public health or safety. However, the lack of a clear separate public interest test has the effect that s 45 may be used to shield from public access documents relating to government contracts via the simple device of including within them confidentiality clauses. The increased resort to commercial in confidence claims has become a significant issue given the continuing trend towards the outsourcing of government services. It should be noted in this regard that documents belonging to private sector contractors are not subject to the Act except to the extent that they are in the possession of the contracting agency. This contrasts with the position in New Zealand where the Official Information Act applies to information held by private contractors performing services for a minister or agency.
A related issue concerns the effect of privatisation in narrowing the range of bodies which are subject to the operation of the Act. A body which is privatised will cease to fall within the definition of an agency if it becomes an incorporated company. Alternatively, it may be added to the list of bodies in Schedule 2 which are excluded in respect of their commercial functions or it may be declared by regulation to be a prescribed body for the purposes of the definition of a 'prescribed authority' in s 4( I).
Apart from the difficulties which flow from the Act's drafting, there is also a clear need for attention to be paid to the implementation of record-keeping practices which have been adapted to take into account the increasing use of electronic communications. Electronic records require new record-keeping standards which deal with issues such as the fact that document formats change and become unreadable, they may not be captured by paper-based records systems, and their context and relationship to other records may be easily lost.
There is. also a need for better administration and resourcing of the Act. Factors which have contributed to its failure to work as intended include a failure to employ sufficient staff to deal in a timely manner with FoI requests, failure to provide adequate training for FoI officers and what the Ombudsman has described as 'a growing void in availability of FoI material for practitioners'. More generally, evidence of poor attitudes by some agencies to FoI which was noted with concern in the Senate Committee's report are reflective of the absence of any body which plays an active role in publicising its existence, monitoring compliance with its provisions and initiating actions to remedy factors which are found to inhibit its effective operation. The Information Law Branch of the Attorney General's Department is currently responsible for the general administration of the FoI Act but the measures taken by it to assist agencies and members of the public are relatively modest.
The high level of usage of the Act for obtaining access to personal records suggests that it has worked well in this regard. However. anecdotal evidence suggests that the figures may be misleading to the extent that personal records which were previously made available informally are now required to be accessed via the Act. One specific area of concern relates to the difficulties that may be encountered by people such as parents or carers who may legitimately wish to access documents containing personal information about a child or a person who otherwise lacks capacity to make a request for access. This arises because of the requirement to consider the effect of disclosure to an applicant as if it is a disclosure to the world at large and the fact that the Act does not allow agencies to impose restrictions on subsequent uses of information accessed. There is also contention concerning the scope of the therapeutic exemption ins 41 (3).This is problematic because it potentially allows agencies to undermine the spirit of FoI by requiring disclosure via a medical practitioner, and because it does not provide any assistance in other situations where disclosure may result in potential harm to an applicant.
The amendment provisions have also worked reasonably well but are subject to two specific deficiencies. The first is that they cannot be utilised except where an applicant has been given lawful access to the documents to which the application for amendment relates. This means that a person who is denied access to a document on the grounds that it is exempt but who is sufficiently aware of its contents to know, or at least, suspect that it is incorrect cannot use the FoI Act to request its amendment. The second problem is that they allow for amendment only on the basis that information is incomplete, incorrect out of date or misleading. They do not therefore fully align with the Privacy Act which requires not only that information should be correct but also that it should be relevant with regard to the purpose for which it was collected.
The 2003 Bill contains what have been described as a 'modest package of measures' that are primarily concerned with improving the existing operation of the Act and the ways in which it interfaces with the Archives and Privacy Acts. A significant feature is the inclusion of a new Part VA which establishes a new office of FoI Commissioner to be located within the Ombudsman's office (Schedule I, cl 68). Other significant measures include a rewording of the objects clause (Schedule I, clause I), reduced time limits for compliance by agencies, a change to the charges regime so that agencies may impose charges only in respect of documents that are released (Schedule I , cls 18-19), and protection against civil liability arising from the bona fide exercise of the discretion to grant access to an exempt document (Schedule I, cl 70).The Bill also addresses
the major deficiencies in amendment provisions by removing the precondition for prior lawful access (Schedule I, cl 30), and including irrelevance as a ground for amendment (Schedule I , cl 29).
Although the Bill removes the entitlement of agencies to refuse an Fol request on workload grounds without having identified any of the documents described (Schedule I, cl 14), it fails to give effect to most of the many other significant changes to the exemption provisions which were suggested in the Open Government report. As a result, it does not operate to make the access regime any less restrictive in its scope.
The most significant changes apply to s 41.The new test for exemption contained in the Bill requires not only that a document should contain personal information but also that its disclosure would constitute a breach of IPP I I and would not, on balance, be in the public interest.There is also a requirement to consider any special relationship between the applicant and the person whose information is contained in a document and a rewording of the cunrent therapeutic exemption so that it is not confined to medical information and simply requires disclosure in a way that minimises detriment (Schedule I, cl 27).
Other changes involve a rewording of s 42 so as to clarify that it requires a dominant purpose test and that it is not applicable where privilege has been waived, and the removal of s 33A as one of the grounds on which an agency may refuse to confirm the existence of a document (Schedule I, cl 28 and 16).
Although the Bill is a very important first step in a long-overdue process of amendment it is arguable that an approach which simply seeks to improve on the existing provisions in the Act fails to address some more fundamental shortcomings.
The first is that the Act is based on a paradigm that fails to address the challenges posed by the increased use of electronic communications or to harness the potential of modem technology to address the fundamental difficulties that have hampered its capacity to serve as a tool for accountability. What is required is the implementation of modem record-keeping standards and the effective use of technology to improve reporting. It also requires a model in which agencies are required to push information out into the public arena rather than waiting for members of the public to make formal requests for access. This would involve a shift from the current time-inefficient and costly 'request and wait' model to one which requires agencies to anticipate requests and use Internet technology to make broad categories of information immediately available in a readily accessible form as is currently required in the United States.
The other fundamental shortcoming is that it is a based on a model which is conceptualised on the basis of traditional democratic theory rather than on broader conceptions of power. Access to information has the potential to empower those who are subject to the exercise of powers which impact on them in their capacity as citizens. However; the ongoing reconfiguration of government means that powers to make decisions concerning the rights and obligations of citizens are no longer confined to what is left of the public sector. This suggests that an approach which results in the definition of the scope of the Act solely on the basis of the formal structure of a government is in need of reconsideration.
There can be no doubt that the Fol Act has been instrumental in eroding the culture of secrecy which once pervaded Australian government and that it has provided an invaluable tool for the exercise of access and amendment rights in respect of personal information. However; it has failed to achieve the level of interaction between the government and the people originally envisaged and there is now a very real danger that it will become redundant in the absence of political will to implement long-overdue reforms.
[*] MOIRA PATERSON teaches law at Monash University. email@example.com
©2004 Moira Paterson (text)
©2004 Stuart Roth (cartoon)
 See generally Stephen Lambie, 'Freedom of 1nfonmat1on, a F1nn1sh clergyman's gift to democracy' (2002) 97 Freedom of Information Review 2.
 The process wh1ch led to its enactment commenced with the establishment of an Interdepartmental Committee to report on any mod1ficat1ons the US FoI Act required for it be appropriate for Australia
 See Freedom of Information Act 1989 (AC1), Freedom of Information Act 1989 (NSW), Information Act 2002 (N1), Freedom of Information Act 1992 (Qld), Freedom of Information Act 1991 (SA), Freedom of Information Act 1991 (Tas), Freedom of Information Act 1982 (Vic)
 Countries which have enacted FoI laws include Canada, Ireland, New Zealand, and the United K1ngdom, which enacted
Freedom of Infonmat1on laws in 1993, 1998, 1982 and 2000, respectively.
 Open Government Report at .
 Commonwealth Ombudsman, Needs to Know Own motion investigation into the administration of the Freedom of Information Act 1982 in Commonwealth agencies June 1999) ( ‘the Needs to Know report')
 See Senate Standing Committee on Legal and Constitutional Affairs, Freedom of Information Act 1982 (1987); ALRGARC, Open Government A Review of the Federal Freedom of Information Act 1982,ALRC Report No 77/ ARC Report No 40 (1995) ( he Open Government report'); Senate Legal And Constrtut1onal Committee, Report of the Inquiry into the Freedom of Information Amendment (Open Government) Bill 2000 (Cth) (5 April 2001) ( he Senate Committee report).
 See Commonwealth Attorney-General Annual Report for 2000/01 2-4. Also the Needs to Know report, above n 6, [3 23]
 See, eg, Paula Walker,'The Western Australian FoI experience 1996-1998 -Any Lessons for Refonn1' ( 1999) 79 FoI Review 7; N Waters, Print Media Use of Freedom of Information Laws in Australia ( 1999)
 See J MacDonald and C H Jones, The Law of Freedom of Information (2003) [9.1 0].
 Canadian Taskforce, Access to Information Making 1t Work for Canada (2000) 9
 MacDonald and Jones, above n 10, [9.08].
 See M Tapscott and N Taylor, 'Few journalists Use the Federal Freedom of Information Act A Study by the Center for Media and Public Policy', <http://www. heritage org/Press/MediaCenter/ FoIA.cfm> at 14 October 2003; McDonald and Jones, above n 10 [9.09].
 For an example of such a mechanism see the Freedom of Information Act 1982 (Vic), s 12
 See the Open Government report, above n 7 [7.7]-[7.8].
 See Robertson v Minister of Pensions ( 1949) 1 KB 227.Formosa v Secretary, Department of Social Security (1988) 81 ALR
687 and discussion in M Allars, Introduction to Australian Administrative Law ( 1990) -[5.102]
 See Commonwealth Attorney-General Annual Report for 2000/01,Table 7.
 See the Open Government report, above n 7 [7 12] See also the Senate Committee Report, above n 7 [3.120].
 5 USC s 552(a)(6)(E)(v)(II)
 See, eg.ALRC, Australia’s Federal Record A Review of the Archives Act 1983 at  See also statistics concerning differences between the amount of fees notified and collected in Commonwealth AttorneyGeneral Annual Report for 2000/01 [1 33], and comments by the Commonwealth Ombudsman 1n Needs to Know report, above n 6 -
 See Freedom of Information (Charges) Regulations (Cth), reg 2(2)
 See, eg News Corporation Ltd v National Companies and Securities Commission  FCA 36; (1984) 52 ALR 277, Kawadias v Commonwealth Ombudsman  FCA 55; (1984) 52 ALR 728 See generally P Bayne and K Rubenstein, 'The Objects of the Freedom of Information (FoI) Acts and the1r Interpretation' (1995) 2 Australian Journal of Administrative Law 114.
 Official Information Act 1982 (NZ), s 5
 See Shergold v Tanner (200 1) 179 ALR ISO wh1ch was upheld on appeal to the High Court in Shergold v Tanner (2002)
 See Esso Australia Resources v Commissioner of Taxation  HCA 67
 See Open Government Report -[10.26].
 See further M Paterson, 'Commercial in Confidence Claims, Freedom of Information and Government Accountability
-A Critique of the ARC's Approach to the Problems Posed by Government Outsourcing' in R Creyke and J McMillan (eds), Administrative justice -the Core and the Fringe (2001) 243
 See G Airo-Farulla, "'Public" and "Private"in Australian Administrative Law' (1992) 3 Pub1ic Law Review 186; C Michle, 'Government by Contract Who is Accountable?'  QUTLawJl 10; (1999) 15 Queensland University of Technology Law Journal 135
 Official Information Act 1982, s 2(5). See also the Irish Freedom of Information Act 1997, s 6(9).
 See generally A Roberts, 'Structural Pluralism and the Right to Information (2001) 5 I Umvers1ty of Toronto Law Journal 243
 M Campbell, 'FoI Access to Electronic Records' ( 1995) 59 FoI Review 70.
 See R Gibbs and J Hazelwood, 'Eiectromc Records -Problem Solved? The Victorian Electronic Records Strategy and the Future of Electronic Record Keeping in Victoria' (2000) <www.provvic.gov.au>.
 Needs to Know report [3.68]
 The Senate Committee report [3.140] See also R Snell, "Administrative Compliance -Evaluating the Effectiveness of Freedom of Information (2001) 93 FoI Review 26.
 See Commonwealth Attorney-General Annual Report for 2000/01, Chapter 4.
 See Wallace v Health Commission of Victoria  VicRp 43; (1985) VR 403 negard1ng the need for applicants to be able to fully understand the nature and significance of making an application for access.
 See the Open Government report n 50 regarding criticisms of another similar requirement.
 See the Open Government report, above n 7 
 The time limits for notification of receipt, notification of decision and the further extension applicable where consultation is required are reduced to 7 days, 21 days and 14 days, respectively· Freedom of Information (Open Government) Bill 2003, Schedule I, cls 10-12
 See Open Government report, Appendix D.
 For a useful discussion of the potential of computer-assisted reporting see S Lambie, 'Computer-assisted reporting and FoI a PhD
research project' (2001) 91 Fol Review 2.
 For a useful overview see M Tankers!"), 'How the Electronic Freedom of Information Act Amendments of 1996
Update Public Access for the Information Age' (1998) Administrative Law Review 421, 422-3 reproduced in R C Vaughn, Freedom of Information (2000) 59, 60-1.
 ALRC, Australia's Federal Record A of the Archives Act 1983, Report No 85  n 155