Discovery And Judicial Review: Have The Goal Posts Moved?

The standard advice I have given clients both contemplating bringing an application for review, and defending one, is that discovery is not generally required. 

There are a number of reasons for that.  First, the incentives are generally on decision-makers to fully disclose the background to their decisions – whether by initial disclosure of documents, or in affidavits.

Secondly, the narrower ambit of the court’s review jurisdiction means that documents are usually less probative than they would be in a commercial case between private parties.

Finally, the ability to request documents under the Official Information Act often goes some way to providing the type of material ordinarily obtained through discovery.

A recent case has caused me to ask whether that approach is unduly conservative.  In New Zealand Steel Limited v Minister of Commerce of Consumer Affairs [2017] NZHC 3232, NZ Steel brought judicial review of a decision by the Minister that steel imports were being subsidised by China to a de-minimis level only, and the subsidisation was not causing material injury to New Zealand markets. New Zealand Steel was concerned that reports from various foreign regulatory investigations had found countervailing subsidisation in the steel sector, and suspected that MBIE may have disregarded key findings in those reports. It also questioned reliance on some evidence.

In its application for review NZ Steel sought discovery of various categories of documents including:

• correspondence relating to MBIE’s investigation, including internal correspondence and correspondence between Government agencies and with foreign Governments (Category 1);
• working documents including drafts of the various reports (Category 2); and
• material considered and/or relied by MBIE in preparing the various reports (Category 3). 

Her Honour Justice Thomas reviewed both the 2012 amendments to the High Court Rules, and the Judicial Review Procedure Act 2016. She concluded that there was no apparent  difference between discovery in judicial review and discovery in an ordinary civil proceeding. It all came down to the issues and to questions of relevance and proportionality.

NZ Steel argued that the relevant test was (following the decision of the Court of Appeal in Air Nelson Limited v Minister of Transport [2008] NZCA 26) whether the report prepared by MBIE for the Minister had given the Minister a fair and accurate picture of what had been in front of MBIE. To determine that, NZ Steel argued, it needed to know what was not in the final report. MBIE by contrast relied on the line of cases which have held that internal deliberations are not discoverable.

Her Honour dismissed the application for discovery in relation to Category 1 and Category 2 material. She found that the Category 1 material would not assist resolution of the pleaded issues. The Category 2 material went to internal deliberative processes of MBIE and under orthodox principles were not discoverable.

However, she concluded that the Category 3 material was relevant and that discovery was proportionate and in the interests of justice. Her Honour could not see the apparent logic in discovery being considered relevant for materials relied on by MBIE in preparing its final report but not  for the earlier reports that were also part of the statutory process. 

So standing back, what does this mean? On the one hand it is easy to dismiss the case as an example of an unusual statutory framework. On the other hand there is some superficial appeal to the argument that a party alleging failure to have regard to relevant considerations “doesn’t know what it doesn’t know” and therefore should be entitled to discovery. If that line of analysis gains any further traction then it would appear likely that applications for discovery will increase although whether or not they are successful remains another matter. This suggests that the costs of potential discovery need to be factored in when advising clients on judicial review options. It might also suggest that a more fulsome approach to initial disclosure in order to head off an application for discovery could be prudent in appropriate cases.