Monday, April 07, 2014

Environmental Reporting Bill: The origins of secrecy

Back in February, the government finally introduced its Environmental Reporting Bill. The bill would require the Ministry for the Environment and Statistics New Zealand to regularly publish reports on the state of our air, water, land, oceans, and climate, including sources of pressure on the environment and the impacts of degradation. Its a good idea, but there's a catch: it includes an odious secrecy clause which forbids disclosure of "information or analysis that will be, or has been, used in an environmental report" without the joint permission of the Secretary for the Environment and the Government Statistician. The clause "applies despite any other enactment", so it effectively ousts the jurisdiction of the OIA (not to mention the BORA).

I was curious about how this clause came to be in the bill, so I sent in an OIA for the details. The Minister's office were very helpful, and made sure that I got a response back in time to submit on the legislation. The formal advice on the bill is on DocumentCloud here (and on GoogleDocs here). It tells an interesting, if tragic, story about how this clause came to be there. Its also a great example of how good intentions can produce toxic outcomes.

First, the good intention: back in August 2013, when Cabinet approved environmental reporting in principle [PDF; see CAB Min (13) 26/6 on p. 24], they:

21. agreed that reporting be undertaken at arm's length from Ministers, according to the Principles and protocols for producers of Tier 1 statistics, in order to provide the public with assurance in the integrity and independence of the data and analysis;

22. noted that this means that Ministers will be given a briefing on the key messages in the days prior to release, but will not see information on the data released in the environmental reports before they are publicly released and will not be able to influence the content of those reports;

[Link added]

Cabinet appeared to think that the government's word on this would be enough, but late during drafting, while ironing out some of the inconsistencies around directing the Government Statistician to report on something (something which prima facie challenged their statutory independence), they decided to give it statutory form. A briefing note to the Minister for the Environment (13-B-02167: Environmental Reporting; starts on p.24 of the document linked above) has the details, and sadly their motivation here was unpleasantly political: to prevent oversight by a future less cooperative PCE:
19. [I]t is important that no party, other than those working on the reports, have access to any of the reports prior to publication. This includes any of the relevant unpublished statistical data to be used in the reports, or to any of the analysis that will make up the reports, as well as the reports themselves. This includes the Parliamentary Commissioner for the Environment and the joint Ministers.

20. Currently the PCE has powers to obtain information under section 19 of the Environment Act 1986. This could include environmental reporting reports, or any information or data relating to the generation of that reporting. While the current PCE has indicated that she has no intention of making such requests prior to publication, this issue could arise with future Commissioners.

21. Section 19(3) of the Environment Act allows persons bound by the provisions of any enactment to maintain secrecy in relation to any matter and not to supply any information to the Commissioner or to produce any document or paper. Our proposed approach is to reflect in the Bill that no party will receive the report and unpublished statistical data to be used in the reports, or any of the analysis that will make up the reports, or the reports themselves, prior to publication. We are working with PCO so that this does not restrict access by either Minister to information required to progress any policy work within their portfolio.

[Link and emphasis added]

A clause forbidding release prior to publication does not seem particularly onerous. So how did that turn into a general requirement for perpetual secrecy? A later Briefing Note ("Background to Drafting of the Environmental Reporting Bill", 13-B-02346; p. 5 of the document linked above) has more information:
17. Provisions have been made in clause 15 of the Bill to ensure reporting is conducted at arm's length from the Government of the day and to similarly restrict PCE's pre-publication access to material and access to other parties. These provisions apply to domain and synthesis reports, including their findings and conclusions.

18. You have advised that you do not wish to limit any provisions of the Official Information Act (OIA). For consistency with restrictions on Ministerial access, provisions to limit pre-release disclosure will also limit the ability of the public to obtain information prior to publication. This is consistent with section 52(3)(b)(ii) of the OIA.

19. Clause 15 also contains a mechanism to ensure that access to information required by Ministers to discharge their other functions (e.g. policy decisions) will not be impeded by this Bill.

20. We believe the clause constitutes a "minor and technical change" within the scope of the authority delegated to you and the Minister for Statistics by Cabinet...

[Link added]

They're right in saying that a general ban on pre-publication release would override the OIA. But then again, its unclear why they even needed to go that far - the OIA already has clauses protecting the relevant interests, and requests could be refused under s18(d) ("that the information requested is or will soon be publicly available"). And its worth noting that Statistics New Zealand manages to protect the security of Tier One statistics without needing this sort of general opt-out (they have statutory secrecy around their survey forms and a secrecy requirement on staff that information only be disclosed in accordance with their official duties). Its also worth noting that the clause as drafted goes well beyond a ban on pre-publication release and instead imposes perpetual secrecy on this material (so e.g. requests for drafts of an environmental report so we can see how it evolved and whether there was any string-pulling could be refused). It is unclear why the clause is so broad when a narrower one will do. Unfortunately, MfE couldn't release their emails and other communications in time (I'm expecting them in another couple of weeks), so we can't see what was driving officials to draft this atrocity.

Finally: that little note about it being a "minor and technical change": firstly, the OIA is quasi-constitutional legislation which establishes and protects a fundamental principle of our government. Overriding it isn't "minor". But that's not the only law this clause overrides - it "applies despite any other enactment". In other words, it overrides the Bill of Rights Act and the Treaty of Waitangi Act - cores of our constitution. And its not exactly hard to see how it could conflict with either (disclosure in an employment case after someone quit alleging manipulation of environmental reporting or in a Treaty claim about environmental damage). PCO's sloppy drafting just took a great, steaming dump not just on freedom of information, but on our entire constitution. Are these people really that clueless?

Update: due to problems with DocumentCloud, I've put the core documents on GoogleDocs here.