Introduction

Since 2006, the Ministry of Business, Innovation and Employment has promulgated rules for government procurement decision making.  The rules have been imposed on government entities by cabinet directions to the affected entities.  Initially, mainly core ministries and departments were bound, but since 2014, when the rules were amended, they have been extended to other government agencies.

The original rules recently featured in a successful judicial review application brought by the Problem Gambling Foundation (PGF) to set aside a Ministry of Health decision to award contracts to the Salvation Army in preference to the PGF, an incumbent, following a request for proposals (RFP) process. The services to be contracted were to cover the final 30-month period of a six-year integrated problem gambling strategy.  The RFP specified eight criteria and associated weightings, each with numerous sub-criteria, and the use of an evaluation panel with a non-voting chair and six voting members. The panel undertook a decision process that included:

• individual pre-scoring;
• group consensus scoring; and
• moderation.

The group consensus discussion resulted in individuals changing their initial scores and the moderation stage, while not changing scorings, produced changes in preferred rankings. The evaluation panel recommended other providers in preference to PGF in all but two minor areas and the ultimate decisions largely followed the panel's recommendations.

From the claim, the High Court judge identified the following key issues:

• whether grounds for review were available;
• whether there had been a failure to follow the evaluation process and criteria set out in the RFP (breach of the mandatory procurement rules and legitimate expectation);
• whether there were material errors justifying review (flawed evaluation methodology); and
• whether the decision had been affected by bias or failure to manage conflicts of interest.

Scope of review

The court distinguished the leading Court of Appeal decision on the reviewability of procurement decisions (Lab Tests (Auckland) v Auckland District Health Board)1, on the basis that the decision- making context was different in the following respects:

• The RFP did not result in a process contract (which Lab Tests had indicated would be an available ground for review);
• The services were required for the purposes of a different enactment;
• The enactment was not prescriptive about process and contained no commercial or budgetary objectives (as the services would be funded by a levy);
• The process was subject to the Mandatory Rules for Procurement by Departments;
• The Mandatory Rules required conflicts of interest to be eliminated and generally supported a broad scope of judicial review;
• The process included no negotiation;
• The subject matter was not complex;
• The proposing parties were not private, profit-seeking firms; and
• There was not the same non-judicial accountability by the decision maker.

The decision also distinguished between the "substantive merits" of a decision and whether it is "logical or reliable or soundly-based".

As a result, the Lab Tests principles, which expressly limit the available scope of judicial review of procurement decisions, did not apply and the decision was open to review on all grounds raised.

Process failures

Unsurprisingly, the court found that departures from the published evaluation process and the Mandatory Rules justified the decision being set aside on one or more bases (ie, error of law, unreasonableness, legitimate expectation and ultra vires – that is, beyond the scope of legal authority). However, the court held that the ground was not restricted to published and mandatory processes – it also encompassed voluntary and internal process rules privately adopted by the panel.

The court found that the panel had materially departed from the published evaluation process, including:

• changing the sub-criteria and their weightings without notice;
• using other non-notified matters in the evaluation; and
• inserting a new evaluation stage (moderation), which used principles not only not notified in, but also in conflict with, the RFP.

The departures breached both the PGF's legitimate expectation regarding process and the requirements of the Mandatory Rules, and were a sufficient basis to set aside the decision. However, the PGF's claims to a legitimate expectation of oral discussion of the proposal and of signalled intentions to an incumbent to change provider were rejected.

Mistake of fact or lack of probative evidence

The court upheld PGF's claim that, because the evaluative methodology was flawed, the results were unreliable. The court held that a mistaken belief by the panel that it was implementing a well- designed process properly was distinct from the evaluation of the substantive merits of the decision. The court accepted the expert evidence of a statistician who was highly critical of the design and implementation of the evaluation process. The substantial variations in individual scores between panel members were indicative that the criteria were poorly defined. Consensus scoring undermined the independent assessment that the RFP required and covered up the problems created by imprecisely defined criteria. The court held that the Ministry of Health lacked the expertise to evaluate competitive tenders of this nature and the skills to ensure a fair and transparent process (which the Mandatory Rules required). These failures were a further basis to set aside the decision.

Conflict of interest or bias

The ministry conceded that six of the seven panel members had disclosed potential conflicts of interest. Two ministry staff were the contract managers for some bidders, one member had a relative who worked for a subcontractor for a bidder and two had worked for bidders (including the PGF) in the past. Because the Mandatory Rules required the elimination of potential conflicts of interest and an evaluation that guaranteed fairness and impartiality, the court held that it was appropriate to apply the apparent bias standard usually reserved for judicial decisions. The panel members had agreed internally that they would exclude all matters of personal knowledge from evaluation and rely solely on the content of the proposals. As the use of personal knowledge was inevitable – it was impossible for panel members to avoid using personal knowledge unconsciously – the evaluation process rules had the overall effect of requiring the judicial standard of impartiality, but the panel composition meant that it could not be met.

The decision marks a major departure from earlier decisions regarding the available scope of judicial review of government procurement decisions. Of particular concern to government entities engaged in procurement will be the numerous bases on which the Lab Tests decision was distinguished.

The court confirmed that cabinet-imposed procurement rules will be strictly applied and that non- complying decisions may be set aside. This outcome is consistent with comments made in the Lab Tests decision. It is unsurprising that the court held that any notified evaluation and selection process (whether derived from the RFP or from published procurement rules) must be followed in the light of those comments, although review of procurement decisions on the basis of unpublished internal rules had not been granted previously.  However, this decision went even further and held that the very existence of cabinet-imposed procurement rules supports the availability of judicial review of procurement decisions on broad grounds.

The Mandatory Rules for Procurement by Departments have since been replaced by the Government Rules of Sourcing (Third Edition 2015).  The conflict of interest provisions of the existing rules now require policies "to identify, notify and manage conflicts of interest" and the exercise of "sound judgment to manage conflicts", rather than their elimination, as the former Mandatory Rules required.

As evaluation panels commonly contain members who are experienced in the specific area of service provision, treating any personal knowledge gained professionally as a form of bias could be a matter of concern for future procurement decisions.  However, it is possible that personal knowledge was treated as bias in this case only as the result of the panel's internal commitment to avoid the use of personal knowledge (which the judge found to be impossible to achieve).

The most controversial aspect of the decision may be the court's willingness to scrutinise the substance of the evaluation decision by critical examination, after the event (and using the actual scoring results), of the design of the notified evaluation process.  If an evaluation panel's recommendation can be challenged on such a basis, many procurement decisions will be susceptible to judicial review challenge.

The decision has been appealed.

A version of this article appeared in the International Law Office newsletter on 20 October 2015