Go-to guy ... Former departmental secretary Dr Allan Hawke, who reviewed the Commonwealth's FOI regime. Photo: Glen McCurtayne
When ministers and senior public servants need a review of a policy or agency, Dr Allan Hawke is their go-to person. He can be relied on to bring a fresh, critical mind, tempered by years of experience and understanding of the workings of government. His recommendations will be sensible without being radical. He was therefore a natural choice for the recent review of the government's information policy, enshrined in the amended Freedom of Information Act and in the Australian Information Commissioner Act. Both acts were passed in 2010 and provided for a review after two years' operation.
Hawke's review was commissioned by then attorney-general Nicola Roxon in October 2012 and his report was tabled by her successor, Mark Dreyfus, on the eve of this year's election campaign. How high a priority it will receive from an incoming attorney-general and government remains to be seen. Judging from the lack of public debate so far, the signs are inauspicious.
Many public servants, particularly in senior positions, sincerely, and rightly, believe it is their professional duty to protect their ministers from political embarrassment.
The review covered the operation of the newly established Office of the Australian Information Commissioner, together with the impact of changes to the Freedom of Information Act 1982, including the abolition of conclusive certificates and the introduction of a new public interest test. It attracted over 80 submissions, mostly from government agencies but also from the media and from FOI experts in the legal profession and in universities.
In general, Hawke finds the new administrative structure to be working well. However, two years is too short a time for all the changes to be bedded down and reasonably assessed, particularly within the tight time allowed for the review. His first recommendation is therefore for another, more thorough review focused on a comprehensive rewriting of the FOI Act, which has been frequently amended and has become unnecessarily complex and awkward to use.
A subsequent review would also allow time for a more considered judgment on the effectiveness of the new, two-tier, external appeal structure, whereby FOI decisions can be reviewed by the OAIC and then by the Administrative Appeals Tribunal. Added to an initial review process conducted within agencies, this allows for three levels of appeal on the merits of FOI decisions, which some commentators, and the OAIC itself, find excessive.
Possible solutions include removing appeal to the AAT altogether or restricting such appeals to matters of law. So far, the number of cases going to the AAT has been significantly reduced, with most appeals being dealt with by the OIAC (over 500 in the last year), as the legislation intended. Hawke recommends that the appeal structure remain unchanged for now, pending further evidence of demand and workload.
In the meantime, a number of minor administrative changes could help reduce the pressure on the OAIC. For example, the FOI Act should be amended to provide the information commissioner with the express power to remit a matter for further consideration by the original decision-maker. The OAIC's formal role in recording and approving extensions of time for the processing of requests should be reduced. The power to review decisions about charges could be delegated to staff other than the three commissioners (the information commissioner, the FOI commissioner and the privacy commissioner).
The structure of fees and charges is another factor affecting the demand for FOI requests and appeals. Here, a difficult balance must be struck between citizens' right of access to publicly owned information and the government's right to recoup some of the costs of servicing FOI requests and discourage frivolous overuse of the process. In the interests of more open government, the new FOI regime abolished the previous application fee of $30 and amended the processing charge of $20 an hour to allow for five hours of free processing, sufficient to deal with straightforward requests. As a consequence, agencies reported a significant increase in requests for non-personal information, which often tend to be more resource-intensive than requests for personal information. (Numbers of requests for personal information, by contrast, have diminished, mainly because some agencies, notably Centrelink, have expanded their practice of releasing personal information without requiring a formal FOI request.)
The review recognises the strains that dealing with FOI places on hard-pressed agencies, particularly in a time of swingeing efficiency dividends. It recommends leaving the initial application and the first five hours without charge, as at present. The next five hours should be charged at a flat rate of $50, after which the hourly rate should rise from $20 to $30, to keep pace with rising staff costs. All other associated costs should also be charged at $30 an hour rather than at the present, varied rates for different types of processing.
Increased fees could also help to reduce the large number of decisions being referred to the information commissioner for review. At present, no fee is needed to lodge an appeal, a factor that, in the view of the OAIC, has discouraged disappointed FOI applicants from first pursuing an internal appeal within the agency concerned and has contributed to the OAIC's heavy workload. Hawke recommends an application fee of $400 for lodging an appeal, about half the fee now payable for an AAT appeal. As with the AAT, this fee would be reducible to $100 in the case of hardship or if the appeal favours the applicant.
Information advocates may baulk at the deliberate introduction of financial barriers to access. As Hawke notes, the current level of appeals could be a temporary spike, caused by the novelty of the new FOI law and appeal procedures. In the meantime, however, given the absence of any political will to increase its administrative resources to meet demand, the OAIC has little choice.
Besides administrative details, the review also considers some major policy issues relating to the scope of FOI legislation. One is the range of organisations to which the FOI regime applies. For example, should the parliamentary departments of the Senate, the House of Representatives and parliamentary services be subject to FOI? The departments were originally excluded on the constitutional principle that FOI applied to the executive only and not to the other branches of government, including the Parliament. However, the Parliamentary Service Act 1999 appeared to bring the departments within the FOI Act's definition of an ''agency''. In response, the Parliament passed a bill this year specifically excluding the parliamentary officers and their departments from FOI, pending the outcome of the Hawke review. The review, in turn, recommends that the FOI Act apply to the parliamentary departments, but for administrative matters only. This approach, supported by the departments themselves, would bring them into line with other branches of government, including courts, tribunals and the official secretary to the governor-general.
The other vexed issue of exclusion concerns the intelligence agencies, such as the Australian Secret Intelligence Service and ASIO. At present these agencies are wholly exempt from the FOI Act. This blanket exemption follows the example of Britain, always a bastion of executive privilege. The United States, by contrast, with a stronger tradition of open government, allows its security organisations to be subject to FOI, as do Canada and New Zealand. To protect disclosure of sensitive material, these countries rely instead on a strong exemption clause for individual matters of national security. This clause applies across the board to all government organisations, but naturally bites more with intelligence agencies.
FOI advocates have consistently argued for the US approach to be applied in Australia, as have the Information Commissioner, Professor John McMillan, and the FOI Commissioner, Dr James Popple. It is certainly more consistent with the principles of FOI, that particular types of information should be kept secret rather than particular types of agency shielded from scrutiny.
In response, the intelligence agencies predictably argue that removing the blanket exemption would risk national security. It could also undermine relationships with key intelligence partners (even though many of these partners, including the US, lack a similar agency-wide exemption). Hawke does not wish to disturb the status quo and comes down on their side.
The case for including intelligence agencies is certainly sound in principle. In practice, however, it might make little difference. In the US, for example, critics complain that very little information about intelligence agencies gets past the security exemptions. Congress has approved a number of extra limits to America's FOI Act. In addition, the courts have been reluctant to challenge the executive's right to define risks to security. They have allowed liberal use of the ''mosaic'' theory in relation to security information. This view holds that individual items of information, though seemingly innocuous and disclosable, can be pieced together to yield important, highly sensitive material and should therefore be kept secret.
Among the other exemptions in the Australian FOI legislation, as Hawke recognises, the most contentious have been those protecting the confidentiality of public service advice to ministers. With the abolition of the former conclusive certificates, this function has now fallen to two specified exemptions: for cabinet documents and for deliberative processes. The former is an unqualified exemption applying to all cabinet documents (though there can be some dispute over what that term covers). The latter is a conditional exemption that requires a balancing, public interest test to determine whether non-disclosure of matters relating to a deliberative process is justifiable.
The two exemptions overlap in scope and Hawke suggests some minor clarifications. The larger issue remains whether the exemptions together, as currently applied, are overprotective of official confidentiality or whether they threaten the working relations between ministers and public servants. Hawke sides with the FOI advocates who disparage the claim that FOI undermines free and frank advice. Public service advice, he argues, should be robust enough to withstand public disclosure and scrutiny.
This response, however, does not meet the greater concern that public service policy advice, when not acted upon, can be used to embarrass ministers politically. The new FOI Act expressly prohibits the avoidance of political embarrassment for the government as a justification for refusing disclosure. Yet many public servants, particularly in senior positions, sincerely, and rightly, believe that it is their professional duty to protect their ministers from political embarrassment. Unsurprisingly, therefore, they have few scruples about stretching the exemptions beyond their intended meaning.
This area, where governments would prefer to conceal their inner disagreements, is the prime focus of news-hungry journalists and political oppositions. The two exemptions (cabinet documents and deliberative processes) will continue to be the focus of significant appeals, even if these exemptions are comparatively rarely used (together, they comprise about 2 per cent of all exemptions applied).
The media's motives in promoting FOI are revealed in their attitude to the growing use of disclosure logs, whereby agencies post all FOI-disclosed information online for general information. These logs are a significant addition to open government, and the review supports a suggestion that they all be collated in a single website: either the OAIC's or on the government's data portal.
Media organisations, however, object that, if they have paid for a request, they should have sole use of the information, at least for a time. In their view, the media's right to a scoop trumps the public's right to know, though they argue that, without their self-interested pursuit of news, the public would end up being less well-informed. The Greens, too, support a short ''grace'' period before information is posted on the log, presumably to preserve the publicity value of their exposé´s. Hawke agreed, recommending a period of five working days before documents are published on a disclosure log.
This issue reveals the distance still to be travelled before disclosable information becomes routinely available. Ideally, the new information policy is intended to encourage a proactive approach to government information, with agencies voluntarily publishing on their websites all non-personal information that is properly disclosable under FOI legislation. In this case, FOI requests would become rare procedures of last resort rather than a standard means of gaining interesting items of news. Journalists would need to construct their own stories from publicly available information, as they currently do with government statistics or the budget papers. They would no longer be able to rely on government resistance to disclosure to safeguard their monopolistic access to news.
Richard Mulgan is an emeritus professor with the Crawford School of Public Policy at the Australian National University.