The Government is reviewing the Official Information Act behind closed doors. Most readers will immediately grasp the irony.
This review of the OIA wasn’t announced. It wasn’t publicised. It came to light because transparency advocate Andrew Ecclestone happened to be told about it, then revealed it to attendees at a parliamentary forum on democracy last month.
Newsroom’s Sam Sachdeva followed up, getting confirmation from Justice Minister Paul Goldsmith that the Ministry of Justice had quietly commissioned consultancy firm TBL to research the cost drivers of OIA responses across 40 public service agencies.
There was no press release. No public consultation. No statement of intent. Instead it’s a review of the law that exists to open government up, conducted in the dark.
This is more than just ironic. It’s also typical of how New Zealand governments — every single one of them — have treated the OIA since it passed in 1982.
But this time the situation is sharper. For decades, the erosion of the OIA was largely a cultural problem: a slow drift toward secrecy driven by bureaucratic incentives and ministerial convenience. What’s changed is that the threat has become explicit and political.
The Government is now openly discussing changing not just how the OIA is administered, but what the legislation covers. Goldsmith’s comments to Newsroom were telling: he’s concerned that under current OIA procedures, “every different little element of communication has been included.” He’s not really talking about admin delays. He’s talking about cutting back what the public gets to see. It’s a signal that the Government may want to narrow what citizens are actually entitled to see.
This is the crossroads. The OIA could be weakened in ways it has never been weakened before, using a manufactured cost narrative as justification.
Or, and this is genuinely possible, the current moment of scrutiny could produce the reform push that the law has needed for years to be improved and updated. There are voices making that case, some of them from unexpected quarters.
Whether that happens now depends on pressure. That’s what this column is for.
The False narrative about costs
The government’s stated rationale for the review of the OIA is a claimed 394% increase in OIA requests since 2016. The number sounds alarming. It shouldn’t be taken at face value.
Ecclestone has gone through this claim in detail. It turns out that the spike is, in large part, a counting artefact. Agencies have been expanding what they classify as an OIA request. Some are now logging insurance company queries about traffic accidents. Others are counting finance companies’ requests to track down debtors. Police have been adding requests from people challenging speeding tickets by seeking speed camera images.
Also, government job cuts mean that departmental experts who once answered informal enquiries off the cuff are no longer employed — or are now required to log those conversations as formal requests.
So, this is not some great democratic wave. A lot of it is just agencies counting differently.
There’s also a more structural explanation. The No Right Turn blog, which has tracked the OIA closely for years, makes the point bluntly: much of the real increase in requests is coming from service delivery agencies — Corrections, ACC, MSD — because the government has become more adversarial. It’s denying benefits, tightening entitlements, cutting services, and so people are using the tools available to them to push back and enforce their rights. Which is exactly how the OIA is supposed to work. Using that as a justification to weaken it is a deeply cynical move.
The Culture problem
Before last week’s news, the sharpest piece on the OIA in recent months was David Harvey’s essay in the Herald in February: “How a transparency law was hollowed out.” Harvey is a former district court judge, and his analysis is damning without being hysterical. The OIA legislation, he argues, is fundamentally sound. The problem is the culture that has grown up around it: strategic delays, political interference, preferential treatment, and the quiet erosion of record-keeping. His closing question was sharp: “If transparency is the law’s default, what is it that they fear?”
The record-keeping point matters more than it first sounds. It’s not just that agencies delay or redact. It’s that officials have, in some cases, been actively discouraged from creating records in the first place.
Health New Zealand has been the most egregious recent offender. The Ombudsman found the agency had been operating practices “contrary to law.” In September, Andrea Vance reported that Health NZ had unlawfully withheld nurse staffing data for a year — data showing 37-40% of hospital shifts were understaffed — while publicly claiming to be over-recruiting nurses.
The PSA’s national secretary Fleur Fitzsimons was pointed about who bore responsibility: “He can’t keep demanding savings and then blame officials when the impacts of cuts are felt.” The minister sets the resourcing. The minister owns the failure. Courts have agreed — in December, the High Court ordered the Corrections chief executive personally to obey the law, raising the prospect of fines or imprisonment for continued non-compliance.
And still, the Government wants to narrow the OIA rather than fix its agencies.
Matthew Hooton observed in 2017, in what has become a kind of dark axiom for OIA watchers, that the trend across governments was clear: “The Muldoon regime administered the OIA well, the Lange government did so only adequately, the Bolger-Shipley government reluctantly, the Clark government disgracefully and the Key-English government abused it shamelessly.” Each new government, he said, was worse than the last. The trendline suggested the worst was always still to come. There’s no reason to think the current government is the exception.
The paywall now starts at halfway through all Democracy Project newsletters. Please take out a paid sub if you want to support this service and access the full content, including the following sections: “Enter AI — for better and worse”; “The Finlayson surprise, and the Allen problem”, “What would actually help”, and “Why this matters in an election year”