Those who deliberately breach the OIA should fear more than being mildly criticised by the ombudsman (Image: Tina Tiller)
Those who deliberately breach the OIA should fear more than being mildly criticised by the ombudsman (Image: Tina Tiller)

OPINIONPoliticsDecember 14, 2021

A few simple steps to improve the OIA system

Those who deliberately breach the OIA should fear more than being mildly criticised by the ombudsman (Image: Tina Tiller)
Those who deliberately breach the OIA should fear more than being mildly criticised by the ombudsman (Image: Tina Tiller)

A retired public servant with three decades of experience suggests a ‘fix’ for the OIA system. And it’s not another agency.

The Official Information Act (OIA) has come a long way in almost 40 years. The scale of its use is quite staggering – nearly 26,000 requests in 2020 to agencies covered by the Public Service Commission. Reliable data for additional agencies (eg, local bodies) escaped my limited search skills, but I’m sure you could add another 10,000 or so from that.

It has become an important part of our democratic landscape – for individuals, organisations, political parties in opposition, and journalists. For the latter, in particular good investigative ones with a keen sense of smell, it can often mean an exclusive scoop. If Peter Hughes, the public service commissioner, is to be taken at face value then it’s also apparently working pretty well – some 97% of requests are answered on time.

That didn’t feel right to me, and some good investigative journalism from Stuff recently confirmed my suspicions – ironically using the OIA itself to prove their point. They found unsurprisingly that the 97% calculation includes those requests answered within the extended time frame specified by the answering agency. So timeliness in those cases was measured against the additional bureaucratically imposed timeframe, regardless of how reasonable it was for the agency to extend it in the first place or how reasonable the length of the extension was.

Certainly my own experience as a requestor is that the 20-working-day time limit is a target, not a deadline, and where it is not met it is brushed off as an inevitable consequence of the nature of the request.

Mr Ingles (a pseudonym) wrote for The Spinoff earlier this year about his insider experience of “delays, obfuscations and inexplicable redactions”. As an insider I witnessed the delays, and it was part of my job to ensure there was no obfuscating, and that redactions (information withheld) were justified. But as an outsider I have experienced all three.

Mr Ingles’ solution to these problems is to create an “independent body” to handle all OIA requests, as I understand it in the name of impartiality.

Be careful what you wish for, Mr Ingles! Such a body would need to be enormous to sweep up the OIA work of every agency. Not only that but it would add an unknown period of additional time to the mix due to seeking access to all the relevant information held by an agency or agencies. How could these new super public servants be assured that they got it all, or even that they asked the right agency?

And how, with no specialist knowledge of the subject matter, could they make an assessment of the information against the withholding principles in the OIA, for example, serious damage to the economy or commercial sensitivity? Just not feasible, in my view.

I agree that there is a problem with the implementation of the act. I have seen it from both sides – as an outsider since 2011, I have had to use the ombudsman complaint process on occasion to achieve a suitable outcome. In one instance, instead of the 17 pages that the agency was prepared to release to me (and that was only after I complained to the ombudsman), after the full ombudsman investigation I got the full file, some 5cm thick. But that took almost a year, and I was someone who understood the processes.

But when it comes to “obfuscation and inexplicable redactions” then politics is often the major culprit. Since the mid 1980s (Roger Douglas was one of the first), ministers have routinely staffed their offices, at the senior level at least, with political appointees rather than career public servants. As well, there has been a proliferation of communications advisers and public relations advisers, better categorised as political risk managers.

More often than not this means there is no meaningful understanding of the OIA within those offices. I experienced one office where the elementary rule was to withhold anything that could be remotely politically embarrassing. Forget about the act!

The two seemingly outlandish examples given by Ingles (emails and aide memoires are not allegedly covered by the OIA) are real life. I have experienced them, and successfully shot them down, both as an insider and an outsider. But it all takes time.

This sort of thing is currently playing out between the department of prime minister and cabinet (DPMC) and Stuff (and inevitably, the ombudsman) as DPMC reportedly withheld briefings for the prime minister on the Groundswell protests, claiming “the documents concerned are not ones that get released” as covered by a “longstanding practice” relating to certain types of briefings to the PM. You’ll be har pressed to find any reference within the OIA to such document types or “longstanding practices” . The act considers the content of documents, not the type of document or some alleged practice. An example of pure ignorance/incompetence? Or something more serious – politics-based decision making? The ombudsman will decide.

Another significant factor in is the sheer volume of information created in the electronic age. Good old physical files which used to contain everything relevant seldom exist now, and electronic filing is haphazard. For example, not all emails on an issue would find their way into a central place.

The final significant factor is simply resources. Often the staff best placed to understand and deal with a request are those still working on the very issue that the request relates to. Their priority is almost always the substantive work. Our department of labour mantra that OIA work was “core, not a chore” used to fall on deaf ears. This goes back to the fact that agencies were never given any extra funding to make the OIA work efficiently.

It’s easy to set out all these problems. But how do we fix them?

Personally I think “fix” is too big an ask, but certainly there are ways of making improvements. The act itself is not so much the problem as the agencies administering it.

The key improvement opportunity already lies within the OIA – a reason for not releasing information is that it’s “already publicly available”. So make more of it available online! It does happen to a limited extent already, but it is haphazard. Apparently agencies have a “requirement” to release cabinet papers on behalf of their ministers, but only when the minister has authorised them for release. Well that’s going to work, isn’t it?

The Public Service Commission says that it “encourages” proactive release. At present some agencies classify publication of requests and releases already made as proactive release of information, when it is obviously reactive. Peter Hughes says he “would like to get more information out there before it is asked for”. Well what are you waiting for?

There is also a great private sector (and voluntary) attempt at exposing more official information. FYI publishes all responses to OIA requests made through the website. While this is an excellent initiative and resource, once again it is haphazard in that not enough people know about the website even fewer use it for their OIA request.

Essentially all this haphazard mix of publicly available information should be better organised and mandated. The ultimate mandate would be by legislation. But it’s not straightforward given the myriad types of information and the multiplicity of agencies. But it could be optimised in a staged way.

First, in a non-legislative context, almost immediate steps could be taken if there was the bureaucratic will to do it, and political support. For example, cabinet could make new meaningful rules about ministerial information. Ministers could include proactive release requirements in their expectation documents for agencies. The Public Service Commission could include such requirements or expectations in employment agreements for chief executives, not to mention in the same documents also making OIA compliance and activity part of an agency’s core business.

Second, an organisation like FYI should be given government funding to expand their reach, both in a publicity and marketing sense and in a capacity and capability sense.

Third, create a general statutory duty on agencies to take all practicable steps to make information publicly available. Support that with a regulated list (i.e, changeable by regulation) of types of information that must be published within a specified time of its creation.

The OIA already starts with a principle of availability. What are we waiting for?

Keep going!